HVAC Water Chillers HVAC Water...HVAC Water Chillers and Cooling Towers Fundamentals, Application,...

HVAC Water Chillersand Cooling TowersFundamentals, Application,and Operation

Herbert W. StanfordStanford White Associates Consulting Engineers, Inc.

Raleigh, North Carolina, U.S.A.

M A R C E L

MARCEL DEKKER, INC. NEW YORK • BASELm

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress

ISBN: 0-8247-0992-6

This book is printed on acid-free paper.

Headquarters

Marcel Dekker, Inc.

270 Madison Avenue, New York, NY 10016

tel: 212-696-9000; fax: 212-685-4540

Eastern Hemisphere Distribution

Marcel Dekker AG

Hutgasse 4, Postfach 812, CH-4001 Basel, Switzerland

tel: 41-61-260-6300; fax: 41-61-260-6333

World Wide Web

http://www.dekker.com

The publisher offers discounts on this book when ordered in bulk quantities. For more

information, write to Special Sales/Professional Marketing at the headquarters address

above.

Copyright q 2003 by Marcel Dekker, Inc. All Rights Reserved.

Neither this book nor any part may be reproduced or transmitted in any form or by any

means, electronic or mechanical, including photocopying, microfilming, and recording, or

by any information storage and retrieval system, without permission in writing from the

publisher.

Current printing (last digit):

10 9 8 7 6 5 4 3 2 1

PRINTED IN THE UNITED STATES OF AMERICA

MECHANICAL ENGINEERINGA Series of Textbooks and Reference Books

Founding EditorL. L. Faulkner

Columbus Division Battelle Memorial Instituteand Department of Mechanical Engineering

The Ohio State UniversityColumbus Ohio

1 Spring Designer's Handbook, Harold Carlson2 Computer-Aided Graphics and Design, Daniel L Ryan3 Lubncation Fundamentals, J George Wills4 Solar Engmeenng for Domestic Buildings, William A Himmelman5 Applied Engmeenng Mechanics Statics and Dynamics, G Boothroyd and

C Poh6 Centnfugal Pump Clinic, Igor J Karassik7 Computer-Aided Kinetics for Machine Design, Daniel L Ryan8 Plastics Products Design Handbook, Part A Matenals and Components, Part

B Processes and Design for Processes, edited by Edward Miller9 Turbomachmery Basic Theory and Applications, Earl Logan, Jr

10 Vibrations of Shells and Plates, Wemer Soedel11 Flat and Corrugated Diaphragm Design Handbook, Mario Di Giovanni12 Practical Stress Analysis in Engineering Design, Alexander Blake13 An Introduction to the Design and Behavior of Bolted Joints, John H

Bickford14 Optimal Engmeenng Design Pnnciples and Applications, James N Siddall15 Spnng Manufacturing Handbook Harold Carlson16 Industnal Noise Control Fundamentals and Applications, edited by Lewis H

Bell17 Gears and Their Vibration A Basic Approach to Understanding Gear Noise,

J Derek Smith18 Chains for Power Transmission and Material Handling Design and Appli-

cations Handbook, American Chain Association19 Corrosion and Corrosion Protection Handbook, edited by Philip A

Schweitzer20 Gear Dnve Systems Design and Application, Peter Lynwander21 Controlling In-Plant Airborne Contaminants Systems Design and Cal-

culations, John D Constance22 CAD/CAM Systems Planning and Implementation, Charles S Knox23 Probabilistic Engmeenng Design Pnnciples and Applications, James N

Siddall24 Traction Dnves Selection and Application, Frederick W Heilich III and

Eugene E Shube25 Finite Element Methods An Introduction, Ronald L Huston and Chris E

Passerello

26. Mechanical Fastening of Plastics: An Engineering Handbook, Brayton Lincoln,Kenneth J. Gomes, and James F. Braden

27. Lubrication in Practice: Second Edition, edited by W. S. Robertson28. Principles of Automated Drafting, Daniel L. Ryan29. Practical Seal Design, edited by Leonard J. Martini30. Engineering Documentation for CAD/CAM Applications, Charles S. Knox31. Design Dimensioning with Computer Graphics Applications, Jerome C.

Lange32. Mechanism Analysis- Simplified Graphical and Analytical Techniques, Lyndon

O. Barton33. CAD/CAM Systems: Justification, Implementation, Productivity Measurement,

Edward J. Preston, George W. Crawford, and Mark E. Coticchia34. Steam Plant Calculations Manual, V. Ganapathy35. Design Assurance for Engineers and Managers, John A. Burgess36. Heat Transfer Fluids and Systems for Process and Energy Applications,

Jasbir Singh37. Potential Flows: Computer Graphic Solutions, Robert H. Kirchhoff38. Computer-Aided Graphics and Design: Second Edition, Daniel L. Ryan39. Electronically Controlled Proportional Valves: Selection and Application,

Michael J. Tonyan, edited by Tobi Goldoftas40. Pressure Gauge Handbook, AMETEK, U.S. Gauge Division, edited by Philip

W. Harland41. Fabric Filtration for Combustion Sources: Fundamentals and Basic Tech-

nology, R. P. Donovan42. Design of Mechanical Joints, Alexander Blake43. CAD/CAM Dictionary, Edward J. Preston, George W. Crawford, and Mark E.

Coticchia44. Machinery Adhesives for Locking, Retaining, and Sealing, Girard S. Haviland45. Couplings and Joints. Design, Selection, and Application, Jon R. Mancuso46. Shaft Alignment Handbook, John Piotrowski47. BASIC Programs for Steam Plant Engineers: Boilers, Combustion, Fluid

Flow, and Heat Transfer, V. Ganapathy48. Solving Mechanical Design Problems with Computer Graphics, Jerome C.

Lange49. Plastics Gearing: Selection and Application, Clifford E. Adams50. Clutches and Brakes: Design and Selection, William C. Orthwein51. Transducers in Mechanical and Electronic Design, Harry L. Trietley52. Metallurgical Applications of Shock-Wave and High-Strain-Rate Phenom-

ena, edited by Lawrence E. Murr, Karl P. Staudhammer, and Marc A.Meyers

53. Magnesium Products Design, Robert S. Busk54. How to Integrate CAD/CAM Systems. Management and Technology, William

D. Engelke55. Cam Design and Manufacture: Second Edition; with cam design software

for the IBM PC and compatibles, disk included, Preben W. Jensen56. Solid-State AC Motor Controls: Selection and Application, Sylvester Campbell57. Fundamentals of Robotics, David D. Ardayfio58. Belt Selection and Application for Engineers, edited by Wallace D. Erickson59. Developing Three-Dimensional CAD Software with the IBM PC, C. Stan Wei60. Organizing Data for CIM Applications, Charles S. Knox, with contributions

by Thomas C. Boos, Ross S Culverhouse, and Paul F. Muchnicki

61. Computer-Aided Simulation in Railway Dynamics, by Rao V. Dukkipati andJoseph R. Amyot

62. Fiber-Reinforced Composites: Materials, Manufacturing, and Design, P. K.Mallick

63. Photoelectric Sensors and Controls Selection and Application, Scott M.Juds

64. Finite Element Analysis with Personal Computers, Edward R. Champion,Jr., and J. Michael Ensminger

65. Ultrasonics: Fundamentals, Technology, Applications: Second Edition,Revised and Expanded, Dale Ensminger

66. Applied Finite Element Modeling: Practical Problem Solving for Engineers,Jeffrey M. Steele

67. Measurement and Instrumentation in Engineering. Principles and BasicLaboratory Experiments, Francis S. Tse and Ivan E. Morse

68. Centrifugal Pump Clinic: Second Edition, Revised and Expanded, Igor JKarassik

69. Practical Stress Analysis in Engineenng Design- Second Edition, Revisedand Expanded, Alexander Blake

70 An Introduction to the Design and Behavior of Bolted Joints. SecondEdition, Revised and Expanded, John H. Bickford

71. High Vacuum Technology. A Practical Guide, Marsbed H. Hablanian72. Pressure Sensors. Selection and Application, Duane Tandeske73. Zinc Handbook' Properties, Processing, and Use in Design, Frank Porter74. Thermal Fatigue of Metals, Andrzej Weronski and Tadeusz Hejwowski75 Classical and Modem Mechanisms for Engineers and Inventors, Preben W.

Jensen76. Handbook of Electronic Package Design, edited by Michael Pecht77. Shock-Wave and High-Strain-Rate Phenomena in Materials, edited by Marc

A. Meyers, Lawrence E. Murr, and Karl P. Staudhammer78. Industrial Refrigeration: Principles, Design and Applications, P. C Koelet79. Applied Combustion, Eugene L. Keating80 Engine Oils and Automotive Lubncation, edited by Wilfried J. Bartz81. Mechanism Analysis: Simplified and Graphical Techniques, Second Edition,

Revised and Expanded, Lyndon O. Barton82. Fundamental Fluid Mechanics for the Practicing Engineer, James W

Murdock83. Fiber-Reinforced Composites: Materials, Manufacturing, and Design, Second

Edition, Revised and Expanded, P. K. Mallick84. Numencal Methods for Engineenng Applications, Edward R. Champion, Jr.85 Turbomachmery. Basic Theory and Applications, Second Edition, Revised

and Expanded, Earl Logan, Jr.86. Vibrations of Shells and Plates: Second Edition, Revised and Expanded,

Wemer Soedel87. Steam Plant Calculations Manual: Second Edition, Revised and Ex panded,

V. Ganapathy88. Industrial Noise Control: Fundamentals and Applications, Second Edition,

Revised and Expanded, Lewis H. Bell and Douglas H. Bell89. Finite Elements- Their Design and Performance, Richard H MacNeal90. Mechanical Properties of Polymers and Composites: Second Edition, Re-

vised and Expanded, Lawrence E. Nielsen and Robert F. Landel91. Mechanical Wear Prediction and Prevention, Raymond G. Bayer

92. Mechanical Power Transmission Components, edited by David W. Southand Jon R. Mancuso

93. Handbook of Turbomachinery, edited by Earl Logan, Jr.94. Engineering Documentation Control Practices and Procedures, Ray E.

Monahan95. Refractory Linings Thermomechanical Design and Applications, Charles A.

Schacht96. Geometric Dimensioning and Tolerancing: Applications and Techniques for

Use in Design, Manufacturing, and Inspection, James D. Meadows97. An Introduction to the Design and Behavior of Bolted Joints: Third Edition,

Revised and Expanded, John H. Bickford98. Shaft Alignment Handbook: Second Edition, Revised and Expanded, John

Piotrowski99. Computer-Aided Design of Polymer-Matrix Composite Structures, edited by

Suong Van Hoa100. Friction Science and Technology, Peter J. Blau101. Introduction to Plastics and Composites: Mechanical Properties and Engi-

neering Applications, Edward Miller102. Practical Fracture Mechanics in Design, Alexander Blake103. Pump Characteristics and Applications, Michael W. Volk104. Optical Principles and Technology for Engineers, James E. Stewart105. Optimizing the Shape of Mechanical Elements and Structures, A. A. Seireg

and Jorge Rodriguez106. Kinematics and Dynamics of Machinery, Vladimir Stejskal and Michael

Valasek107. Shaft Seals for Dynamic Applications, Les Horve108. Reliability-Based Mechanical Design, edited by Thomas A. Cruse109. Mechanical Fastening, Joining, and Assembly, James A. Speck110. Turbomachinery Fluid Dynamics and Heat Transfer, edited by Chunill Hah111. High-Vacuum Technology: A Practical Guide, Second Edition, Revised and

Expanded, Marsbed H. Hablanian112. Geometric Dimensioning and Tolerancing: Workbook and Answerbook,

James D. Meadows113. Handbook of Materials Selection for Engineering Applications, edited by G.

T. Murray114. Handbook of Thermoplastic Piping System Design, Thomas Sixsmith and

Reinhard Hanselka115. Practical Guide to Finite Elements: A Solid Mechanics Approach, Steven M.

Lepi116. Applied Computational Fluid Dynamics, edited by Vijay K. Garg117. Fluid Sealing Technology, Heinz K. Muller and Bernard S. Nau118. Friction and Lubrication in Mechanical Design, A. A. Seireg119. Influence Functions and Matrices, Yuri A. Melnikov120. Mechanical Analysis of Electronic Packaging Systems, Stephen A.

McKeown121. Couplings and Joints: Design, Selection, and Application, Second Edition,

Revised and Expanded, Jon R. Mancuso122. Thermodynamics: Processes and Applications, Earl Logan, Jr.123. Gear Noise and Vibration, J. Derek Smith124. Practical Fluid Mechanics for Engineering Applications, John J. Bloomer125. Handbook of Hydraulic Fluid Technology, edited by George E. Totten126. Heat Exchanger Design Handbook, T. Kuppan

127. Designing for Product Sound Quality, Richard H. Lyon128. Probability Applications in Mechanical Design, Franklin E. Fisher and Joy R.

Fisher129. Nickel Alloys, edited by Ulnch Heubner130 Rotating Machinery Vibration: Problem Analysis and Troubleshooting,

Maurice L. Adams, Jr.131. Formulas for Dynamic Analysis, Ronald L. Huston and C. Q. Liu132. Handbook of Machinery Dynamics, Lynn L. Faulkner and Earl Logan, Jr.133. Rapid Prototyping Technology. Selection and Application, Kenneth G.

Cooper134. Reciprocating Machinery Dynamics Design and Analysis, Abdulla S.

Rangwala135. Maintenance Excellence: Optimizing Equipment Life-Cycle Decisions, edi-

ted by John D. Campbell and Andrew K. S. Jardine136. Practical Guide to Industrial Boiler Systems, Ralph L. Vandagriff137. Lubrication Fundamentals' Second Edition, Revised and Expanded, D. M.

Pirro and A. A. Wessol138 Mechanical Life Cycle Handbook: Good Environmental Design and Manu-

facturing, edited by Mahendra S. Hundal139. Micromachining of Engineering Matenals, edited by Joseph McGeough140. Control Strategies for Dynamic Systems. Design and Implementation, John

H. Lumkes, Jr.141. Practical Guide to Pressure Vessel Manufacturing, Sunil Pullarcot142. Nondestructive Evaluation- Theory, Techniques, and Applications, edited by

Peter J.Shull143. Diesel Engine Engineering: Thermodynamics, Dynamics, Design, and

Control, Andrei Makartchouk144 Handbook of Machine Tool Analysis, loan D Marinescu, Constantin Ispas,

and Dan Boboc145. Implementing Concurrent Engineenng in Small Companies, Susan Carlson

Skalak146. Practical Guide to the Packaging of Electronics: Thermal and Mechanical

Design and Analysis, Ali Jamnia147. Bearing Design in Machinery Engineering Tribology and Lubrication,

Avraham Hamoy148 Mechanical Reliability Improvement Probability and Statistics for Experi-

mental Testing, R E Little149. Industnal Boilers and Heat Recovery Steam Generators: Design, Ap-

plications, and Calculations, V. Ganapathy150. The CAD Guidebook. A Basic Manual for Understanding and Improving

Computer-Aided Design, Stephen J. Schoonmaker151. Industnal Noise Control and Acoustics, Randall F. Barren152. Mechanical Properties of Engineered Materials, Wole Soboyejo153. Reliability Verification, Testing, and Analysis in Engineenng Design, Gary S.

Wasserman154. Fundamental Mechanics of Fluids. Third Edition, I. G. Cume155. Intermediate Heat Transfer, Kau-Fui Vincent Wong156. HVAC Water Chillers and Cooling Towers- Fundamentals, Application, and

Operation, Herbert W. Stanford III

Additional Volumes in Preparation

Handbook of Turbomachinery Second Edition, Revised and Expanded,Earl Logan, Jr., and Ramendra Roy

Progressing Cavity Pumps, Downhole Pumps, and Mudmotors, Lev Nelik

Gear Noise and Vibration: Second Edition, Revised and Expanded, J.Derek Smith

Piping and Pipeline Engineering: Design, Construction, Maintenance,Integrity, and Repair, George A. Antaki

Turbomachinery: Design and Theory. Rama S. Gorla and Aijaz AhmedKhan

Mechanical Engineering Software

Spring Design with an IBM PC, Al Dietrich

Mechanical Design Failure Analysis. With Failure Analysis System Softwarefor the IBM PC, David G. Ullman

Preface

There are two fundamental types of HVAC systems designed to satisfy building

cooling requirements: direct expansion (DX) systems, in which there is direct

heat exchange between the building air and the refrigerant, and secondary

refrigerant systems that utilize chilled water as an intermediate heat exchange

medium to transfer heat from the building air to the refrigerant.

Chilled water systems are the heart of central HVAC cooling, providing

cooling throughout a building or group of buildings from one source. Centralized

cooling offers numerous operating, reliability, and efficiency advantages over

individual DX systems and, on a life-cycle basis, can have significantly lower

total cost.

Every central HVAC cooling system is made up of one or more

refrigeration machines, or water chillers, designed to collect excess heat from

buildings and reject that heat to the outdoor air. The water chiller may use the

vapor compression refrigeration cycle or the absorption refrigeration cycle.

Vapor compression refrigeration compressors may be of the reciprocating,

helical screw, or centrifugal type with electric or gas-fired engine prime movers.

The heat collected by the water chiller must be rejected to the atmosphere. This

waste heat can be rejected by air-cooling, in a process that transfers heat directly

from the refrigerant to the ambient air, or by water-cooling, a process that uses

water to collect the heat from the refrigerant and then to reject that heat to

iii

the atmosphere. Water-cooled systems offer advantages over air-cooled systems,

including smaller physical size, longer life, and higher operating efficiency. The

success of their operation depends, however, on the proper sizing, selection,

application, operation, and maintenance of the cooling tower.

Thus, the goal of this book is to provide the HVAC designer, the building

owner and his operating and maintenance staff, the architect, and the mechanical

contractor with definitive and practical information and guidance relative to the

application, design, purchase, operation, and maintenance of water chillers and

cooling towers. The first half of the book discusses water chillers and the second

half addresses cooling towers.

Each of these two topics is treated in separate sections, each of which is

divided into three basic parts:

Fundamentals (Part I) presents the basic information about systems and

equipment. How they work and their various components are described and

discussed.

In Design and Application (Part II), equipment sizing, selection, and

application are discussed. In addition, the details of piping, control, and water

treatment are presented. Finally, special considerations such as noise control,

electrical service, fire protection, and energy efficiency are examined.

Finally, Operations and Maintenance (Part III) takes water chillers and

cooling towers from commissioning through routine maintenance. Chapters on

purchasing equipment include guidelines and recommended specifications for

procurement.

This is not an academic textbook, but a book designed to be useful on a day-

to-day basis and provide answers about water chiller and cooling tower use,

application, and problems. Extensive checklists, design and troubleshooting

guidelines, and reference data are provided.

Herbert W. Stanford III

iv Preface

Contents

Preface iii

WATER CHILLERS

Part 1 Fundamentals

1. Refrigeration Machines 1

2. Chiller Configurations 24

Part II Design and Application

3. Chilled Water System Elements 45

4. Chiller Controls 62

5. Thermal Storage 71

6. Special Chiller Considerations 86

Part III Operation and Maintenance

7. Chiller Operation and Maintenance 95

8. Buying a Chiller 109

v

COOLING TOWERS

Part IV Fundamentals

9. Cooling Tower Fundamentals 115

10. Cooling Tower Components 129

Part V Design and Application

11. Tower Configuration and Application 147

12. Cooling Tower Controls 179

13. Condenser Water Treatment 191

14. Special Tower Considerations 212

Part VI Operation and Maintenance

15. Cooling Tower Operation and Maintenance 225

16. Buying a Cooling Tower 238

17. In-Situ Tower Performance Testing 247

Appendices

Appendix A. Design Ambient Wet Bulb Temperatures 254

Appendix B. Draft Specifications 257

Appendix C. References and Resources 284

Index 289

Contentsvi

1

Refrigeration Machines

1.1. REFRIGERATION BASICS1.1.1. Vapor Compression Refrigeration Cycle

The term refrigeration, as part of a building HVAC system, generally refers to a

vapor compression system wherein a chemical substance alternately changes

from liquid to gas (evaporating, thereby absorbing heat and providing a cooling

effect) and from gas to liquid (condensing, thereby releasing heat). This “cycle”

actually consists of four steps:

1. Compression: Low-pressure refrigerant gas is compressed, thus raising

its pressure by expending mechanical energy. There is a corresponding

increase in temperature along with the increased pressure.

2. Condensation: The high-pressure, high-temperature gas is cooled by

outdoor air or water that serves as a “heat sink” and condenses to a

liquid form at high pressure.

3. Expansion: The high-pressure liquid flows through an orifice in the

expansion valve, thus reducing the pressure. A small portion of the

liquid “flashes” to gas due to the pressure reduction.

4. Evaporation: The low-pressure liquid absorbs heat from indoor air or

water and evaporates to a gas or vapor form. The low-pressure vapor

flows to the compressor and the process repeats.

1

As shown in Figure 1.1, the vapor compression refrigeration system consists of

four components that perform the four steps of the refrigeration cycle. The

compressor raises the pressure of the initially low-pressure refrigerant gas. The

condenser is a heat exchanger that cools the high-pressure gas so that it changes

phase to liquid. The expansion valve controls the pressure ratio, and thus flow

rate, between the high- and low-pressure regions of the system. The evaporator is

a heat exchanger that heats the low-pressure liquid, causing it to change phase

from liquid to vapor (gas).

Thermodynamically, the most common representation of the basic

refrigeration cycle is made utilizing a pressure–enthalpy chart as shown in

Figure 1.2. For each refrigerant, the phase-change line represents the conditions

of pressure and total heat content (enthalpy) at which it changes from liquid to gas

and vice versa. Thus each of the steps of the vapor compression cycle can easily

be plotted to demonstrate the actual thermodynamic processes at work, as shown

in Figure 1.3.

Point 1 represents the conditions entering the compressor. Compression of

the gas raises its pressure from P1 to P2. Thus the “work” that is done by the

compressor adds heat to the refrigerant, raising its temperature and slightly

increasing its heat content. Point 2 represents the condition of the refrigerant

FIGURE 1.1. Basic components of the vapor compression refrigeration system.

Chapter 12

leaving the compressor and entering the condenser. In the condenser, the gas is

cooled, reducing its enthalpy from h2 to h3.

Point 3 to point 4 represents the pressure reduction that occurs in the

expansion process. Due to a small percentage of the liquid evaporating because of

the pressure reduction, the temperature and enthalpy of the remaining liquid is also

reduced slightly. Point 4, then, represents the condition entering the evaporator.

Point 4 to point 1 represents the heat gain by the liquid, increasing its enthalpy

from h4 to h1, completed by the phase change from liquid to gas at point 1.

FIGURE 1.2. Basic refrigerant pressure–enthalpy relationship.

FIGURE 1.3. Ideal refrigeration cycle imposed over pressure–enthalpy chart.

Refrigeration Machines 3

For any refrigerant whose properties are known, a pressure-enthalpy chart

can be constructed and the performance of a vapor compression cycle analyzed

by establishing the high and low pressures for the system. (Note that Figure 1.3

represents an “ideal” cycle and in actual practice there are various departures

dictated by second-law inefficiencies.)

1.1.2. Refrigerants

Any substance that absorbs heat may be termed a refrigerant. Secondary

refrigerants, such as water or brine, absorb heat but do not undergo a phase

change in the process. Primary refrigerants, then, are those substances that

possess the chemical, physical, and thermodynamic properties that permit their

efficient use in the typical vapor compression cycle.

In the vapor compression cycle, a refrigerant must satisfy several (and

sometimes conflicting) requirements:

1. The refrigerant must be chemically stable in both the liquid and vapor

state.

2. Refrigerants for HVAC applications must be nonflammable and have

low toxicity.

3. Finally, the thermodynamic properties of the refrigerant must meet the

temperature and pressure ranges required for the application.

Early refrigerants, developed in the 1920s and 1930s, used in HVAC applications

were predominately chemical compounds made up of chlorofluorocarbons

(CFCs) such as R-11, R-12, and R-503. While stable and efficient in the range of

temperatures and pressures required for HVAC use, any escaped refrigerant gas

was found to be long-lived in the atmosphere. In the lower atmosphere, the CFC

molecules absorb infrared radiation and, thus, contribute to atmospheric

warming. Once in the upper atmosphere, the CFC molecule breaks down to

release chlorine that destroys ozone and, consequently, damages the atmospheric

ozone layer that protects the earth from excess UV radiation.

The manufacture of CFC refrigerants in the United States and most other

industrialized nations was eliminated by international agreement in 1996. While

there is still refrigeration equipment in use utilizing CFC refrigerants, no new

equipment using these refrigerants is now available in the United States.

Researchers found that by modifying the chemical compound of CFCs by

substituting a hydrogen atom for one or more of the chlorine or fluorine atoms

resulted in a significant reduction in the life of the molecule and, thus, reduced the

negative environmental impact it may have. These new compounds, called

hydrochlorofluorocarbons (HCFCs), are currently used in HVAC refrigeration

systems as R-22 and R-123.

While HCFCs have reduced the potential environmental damage by

refrigerants released into the atmosphere, the potential for damage has not been

Chapter 14

totally eliminated. Again, under international agreement, this class of refrigerants

is slated for phaseout for new equipment installations by 2010–2020, with

total halt to manufacturing and importing mandated by 2030, as summarized in

Table 1.1.

A third class of refrigerants, called hydrofluorocarbons (HFCs), are not

regulated by international treaty and are considered, at least for the interim, to be

the most environmentally benign compounds and are now widely used in HVAC

refrigeration systems.

HVAC refrigeration equipment is currently undergoing transition in the use

of refrigerants. R-22 has been the workhorse for positive displacement

compressor systems in HVAC applications. The leading replacements for R-22

in new equipment are R-410A and, to a lesser extent, R-407C, both of which are

blends of HFC compounds.

R-134A, an HFC refrigerant, is utilized in positive-pressure compressor

systems. At least one manufacturer continues to offer negative-pressure

centrifugal compressor water chillers using R-123 (an HCFC), but it is

anticipated that, by 2010, the manufacture of new negative pressure chillers using

HCFCs will be terminated. These same manufacturers already market a positive-

pressure centrifugal compressor systems using R-134A (though one manufac-

turer currently limits sales to outside of the United States since many countries,

particularly in Europe, have accelerated the phaseout of HCFCs).

TABLE 1.1. Implementation of HCFC Refrigerant Phaseout in the United States

Year to be

implemented Clean air act regulations

2010 No production and no importing of HCFC R-22 except for use in

equipment manufactured prior to January 1, 2010. (Consequently,

there will be no production or importing of new refrigeration

equipment using R-22. Existing equipment must depend on stockpiles

or recycling for refrigerant supplies)

2015 No production and no importing of any HCFC refrigerants except for

use in equipment manufactured before January 1, 2020

2020 No production or importing of HCFC R-22. (Since this is the cutoff

date for new equipment using HCFC refrigerants other than R-22,

this should end the installation of new chillers using R-123)

2030 No production or importing of any HCFC refrigerant. (While it is

anticipated that the vast majority of packaged equipment using R-22

will have been replaced by this date, there will still be a significant

number of water chillers using R-123 still in operation. These chillers

must depend on stockpiles or recycling for refrigerant supplies)


Based on the average 20–25 year life for a water chiller (see Chap. 8) and

the HCFC refrigerant phaseout schedule summarized in Table 1.1, owners should

avoid purchasing any new chiller using R-22. After 2005, owners should avoid

purchasing new chillers using R-123.

ASHRAE Standard 34-1989 classifies refrigerants according to their

toxicity (A ¼ nontoxic and B ¼ evidence of toxicity identified) and flammability

(1 ¼ no flame progation, 2 ¼ low flammability, and 3 ¼ high flammability).

Thus, all refrigerants fall within one of the six “safety groups,” A1, A2, A3, B1,

B2, or B3. For HVAC refrigeration systems, only A1 refrigerants should be

considered. Table 1.2 lists the safety group classifications for refrigerants

commonly used in HVAC applications.

1.2. CHILLED WATER FOR HVAC APPLICATIONS

The basic vapor compression cycle, when applied directly to the job of building

cooling, is referred to as a direct-expansion (DX) refrigeration system. This

reference comes from the fact that the building indoor air that is to be cooled

passes “directly” over the refrigerant evaporator without a secondary refrigerant

being utilized. While these cooling systems are widely use in residential,

commercial, and industrial applications, they have application, capacity, and

performance limitations that reduce their use in larger, more complex HVAC

applications. For these applications, the use of chilled water systems is dictated.

Typical applications for chilled water systems include large buildings (offices,

laboratories, etc.) or multibuilding campuses where it is desirable to provide

cooling from a central facility.

As shown in Figure 1.4, the typical water-cooled HVAC system has three

heat transfer loops:

Loop 1 Cold air is distributed by one or more air-handling units to the

spaces within the building. Sensible heat gains, including heat from

TABLE 1.2. HVAC Refrigerant Safety

Groups

Refrigerant Type Safety group

R-22 HCFC A1

R-123 HCFC B1

R-134A HFC A1

R-407C HFC Blend A1

R-410A HFC Blend A1

R-717 Ammonia B2

R-718 Water A1

Chapter 16

temperature-driven transmission through the building envelope; direct

solar radiation through windows; infiltration; and internal heat from

people, lights, and equipment, are “absorbed” by the cold air, raising its

temperature. Latent heat gains, moisture added to the space by air

infiltration, people, and equipment, is also absorbed by the cold air, raising

its specific humidity. The resulting space temperature and humidity

condition is an exact balance between the sensible and latent heat gains

and capability of the entering cold air to absorb those heat gains.

Loop 2 The distributed air is returned to the air handling unit, mixed with

the required quantity of outdoor air for ventilation, and then directed

over the cooling coil where chilled water is used to extract heat from the

air, reducing both its temperature and moisture content so it can be

distributed once again to the space.

As the chilled water passes through the cooling coil in counter flow to the

air, the heat extraction process results in increased water temperature.

The chilled water temperature leaving the cooling coil (chilled water

return) will be 8–168F warmer than the entering water temperature

FIGURE 1.4. Water-cooled HVAC system schematic.


(chilled water supply) at design load. This temperature difference

(range ) establishes the flow requirement via the relationship shown in

Eq. 1.1.

Fchw ¼ Q=ð500 £ RangeÞ ð1:1Þwhere Fchw ¼ chilled water flow rate (gpm), Q ¼ total cooling system

load (Btu/hr), Range ¼ chilled water temperature rise (8F), 500 ¼conversion factor (Btumin/gal8F hr) (1 Btu/lb8 F £ 8.34 lb/gal £ 60

min/hr).

The warmer return chilled water enters the water chiller where it is cooled

to the desired chilled water supply temperature by transferring the heat

extracted from the building spaces to a primary refrigerant. This process,

obviously, is not “free” since the compressor must do work on the

refrigerant for cooling to occur and, thus, must consume energy in the

process. Since most chillers are refrigerant-cooled, the compressor

energy, in the form of heat, is added to the building heat and both must be

rejected through the condenser.

Loop 3 The amount of heat that is added by the compressor depends on

the efficiency of the compressor. This heat of compression must then be

added to the heat load on the chilled water loop to establish the amount

of heat that must be rejected by the condenser to a heat sink, typically the

outdoor air.

1.2.1. Determining Chilled Water SupplyTemperature

The first step in evaluating a chilled water system is to determine the required

chilled water supply temperature.

For any HVAC system to provide simultaneous control of space

temperature and humidity, the supply air temperature must be low enough to

simultaneously satisfy both the sensible and latent cooling loads imposed.

Sensible cooling is the term used to describe the process of decreasing the

temperature of air without changing the moisture content of the air. However, if

moisture is added to the room by the occupants, infiltrated outdoor air, internal

processes, etc., the supply air must be cooled below its dew point to remove this

excess moisture by condensation. The amount of heat removed with the change in

moisture content is called latent cooling. The sum of the two represents the total

cooling load imposed by a building space on the chilled water cooling coil.

The required temperature of the supply air is dictated by two factors:

1. The desired space temperature and humidity setpoint and

2. The sensible heat ratio (SHR) defined by dividing the sensible cooling

load by the total cooling load.

Chapter 18

On a psychrometric chart, the desired space conditions represents one end

point of a line connecting the cooling coil supply air conditions and the space

conditions. The slope of this line is defined by the SHR. An SHR of 1.0

indicates that the line has no slope since there is no latent cooling. The typical

SHR in comfort HVAC applications will range from about 0.85 in spaces with

a large number of people to approximately 0.95 for the typical office.

The intersection between this “room” line and the saturation line on the

psychrometric chart represents the required apparatus dewpoint (ADP)

temperature for the cooling coil. However, since no cooling coil is 100%

efficient, the air leaving the coil will not be at a saturated condition, but will have

a temperature about 1–28F above the ADP temperature.

While coil efficiencies as high as 98% can be obtained, the economical

approach is to select a coil for about 95% efficiency, which typically results in the

supply air wet bulb temperature being about 18F lower than the supply air dry

bulb temperature.

Based on these typical coil conditions, the required supply air

temperature can determined by plotting the room conditions point and a line

having a slope equal to the SHR passing through the room point, determining

the ADP temperature intersection point, and then selecting a supply air

condition on this line based on a 95% coil efficiency. Table 1.3 summarizes

the results of this analysis for several different typical HVAC room design

conditions and SHRs.

For a chilled water cooling coil, approach is defined as the temperature

difference between the entering chilled water and the leaving (supply) air.

While this approach can range as low as 38F to as high as 108F, the typical

value for HVAC applications is approximately 78F. Therefore, to define the

required chilled water supply temperature, it is only necessary to subtract 78Ffrom the supply air dry bulb temperature determined from Table 1.3.

TABLE 1.3. Typical Supply Air Temperature Required to Maintain

Desired Space Temperature and Humidity Conditions

Space conditions

Supply air DB/WB temperature

required

Temperature Relative humidity 0.90þ SHR 0.80–0.89 SHR

75 50% 54/53 52/51

70 50% 50/49 44/43

65 50% 44/43 41/40


1.2.2. Establishing the Temperature Range

Once the required chilled water supply temperature is determined, the desired

temperature range must be established.

From Eq. 1.1, the required chilled water flow rate is dictated by the imposed

cooling load and the selected temperature range. The larger the range, the lower

the flow rate and, thus, the less energy consumed for transport of chilled water

through the system. However, if the range is too large, chilled water coils and

other heat exchangers in the system require increased heat transfer surface and, in

some cases, the ability to satisfy latent cooling loads is reduced.

Historically, a 108F range has been used for chilled water systems, resulting

in a required flow rate of 2.4 gpm/ton of imposed cooling load. For smaller

systems with relatively short piping runs, this range and flow rate are acceptable.

However, as systems get larger and piping runs get longer, the use of higher

ranges will reduce pumping energy requirements. Also, lower flow rates can also

result in economies in piping installation costs since smaller sized piping may be

used.

At a 128F range, the flow rate is reduced to 2.0 gpm/ton and, at a 148Frange, to 1.7 gpm/ton. For very large campus systems, a range as great as 168F(1.5 gpm/ton) to 208F (1.2 gpm/ton) may be used.

1.3. VAPOR COMPRESSION CYCLE CHILLERS

As introduced in Section 1.1, a secondary refrigerant is a substance that does not

change phase as it absorbs heat. The most common secondary refrigerant is water

and chilled water is used extensively in larger commercial, institutional, and

industrial facilities to make cooling available over a large area without

introducing a plethora of individual compressor systems. Chilled water has the

advantage that fully modulating control can be applied and, thus, closer

temperature tolerances can be maintained under almost any load condition.

For very low temperature applications, such as ice rinks, an antifreeze

component, most often ethylene or propylene glycol, is mixed with the water and

the term brine (left over from the days when salt was used as antifreeze) is used to

describe the secondary refrigerant.

In the HVAC industry, the refrigeration machine that produces chilled

water is generally referred to as a chiller and consists of the compressor(s),

evaporator, and condenser, all packaged as a single unit. The condensing medium

may be water or outdoor air.

The evaporator, called the cooler, consists of a shell-and-tube heat

exchanger with refrigerant in the shell and water in the tubes. Coolers are

designed for 3–11 fps water velocities when the chilled water flow rate is

selected for a 10–208F range.

Chapter 110

For air-cooled chillers, the condenser consists of an air-to-refrigerant heat

exchanger and fans to provide the proper flow rate of outdoor air to transfer the

heat rejected by the refrigerant.

For water-cooled chillers, the condenser is a second shell-and-tube heat

exchanger with refrigerant in the shell and condenser water in the tubes.

Condenser water is typically supplied at 70–858F and the flow rate is selected for

a 10–158F range. A cooling tower is typically utilized to provide condenser water

cooling, but other cool water sources such as wells, ponds, and so on, can be used.

1.3.1. Positive Displacement Compressors

Water chillers up to about 100 tons capacity typically utilize one or more positive

displacement type reciprocating compressors.

The reciprocating compressor uses pistons in cylinders to compress the

refrigerant gas. Basically, it works much like a 2-cycle engine except that the

compressor consumes shaft energy rather than producing it. Refrigerant gas

enters the cylinder through an intake valve on the downward stroke of the piston.

The intake valve closes as the piston starts its upward compression stroke, and

when the pressure is high enough to overcome the spring resistance, the discharge

valve opens and the gas leaves the cylinder. The discharge valve closes as the

piston reaches top-dead-center and the cycle repeats itself as the piston starts

down with another intake stroke. The pistons are connected to an offset lobed

crankshaft via connecting rods. The compressor motor rotates the crankshaft, and

this rotational motion is transformed to a reciprocating motion for the pistons.

Control of the reciprocating compressor refrigeration system is fairly

simple. At the compressor, a head-pressure controller senses the compressor

discharge pressure and opens the unloaders on the compressor if this pressure

rises above the setpoint. The unloader is a simple valve that relieves refrigerant

gas from the high-pressure discharge side of the compressor into the low-pressure

suction side, thus effectively raising the inlet pressure and reducing the net

pressure difference that is required of the compressor. The high-pressure setpoint

is based on the condensing requirements and is normally a pressure

corresponding to approximately 1058F for the refrigerant, (R-22 or R-410A).

A temperature sensor located on the suction line leaving the evaporator

modulates the expansion valve to maintain the setpoint. Thus, as the load on the

evaporator changes, the flow rate through the expansion valve is changed

correspondingly. The expansion valve sensor will detect an increased

temperature (i.e., superheat) if the flow rate is too low and a decreased

temperature (i.e., subcooling) if the flow rate is too high. This temperature

setpoint is typically 408F for comfort applications.

Reciprocating water chillers larger than about 20 tons capacity are almost

always multiple-compressor units. In the selection of a multiple compressor


chiller, it is important that the compressors have independent refrigerant circuits

so that in the event of one compressor failing, the remaining one(s) can continue

to operate. Some lower-cost units will have all the compressors operating in

parallel on one refrigerant circuit.

1.3.2. Rotary Compressors

For larger capacities (100 tons to over 10,000 tons), rotary compressor water

chillers are utilized. There are two types of rotary compressors applied: positive

displacement rotary screw compressors and centrifugal compressors.

Figure 1.6 illustrates the rotary helical screw compressor operation. Screw

compressors utilize double mating helically grooved rotors with “male” lobes and

“female” flutes or gullies within a stationary housing. Compression is obtained by

direct volume reduction with pure rotary motion. As the rotors begin to unmesh, a

void is created on both the male and the female sides, allowing refrigerant gas to

flow into the compressor. Further rotation starts the meshing of another male lobe

with a female flute, reducing the occupied volume, and compressing the trapped

gas. At a point determined by the design volume ratio, the discharge port is

uncovered and the gas is released to the condenser.

Capacity control of screw compressors is typically accomplished by

opening and closing a slide valve on the compressor suction to throttle the flow

rate of refrigerant gas into the compressor. Speed control can also be used to

control capacity.

The design of a centrifugal compressor for refrigeration duty originated

with Dr. Willis Carrier just after World War I. The centrifugal compressor raises

the pressure of the gas by increasing its kinetic energy. The kinetic energy is

converted to static pressure when the refrigerant gas leaves the compressor and

expands into the condenser. Figure 1.5 illustrates a typical centrifugal water

chiller configuration. The compressor and motor are sealed within a single

casing and refrigerant gas is utilized to cool the motor windings during

operation. Low-pressure gas flows from the cooler to the compressor. The gas

flow rate is controlled by a set of preswirl inlet vanes that regulate the refrigerant

gas flow rate to the compressor in response to the cooling load imposed on the

chiller.

Normally, the output of the chiller is fully variable within the range 15–

100% of full-load capacity. The high-pressure gas is released into the condenser,

where water absorbs the heat and the gas changes phase to liquid. The liquid, in

turn, flows into the cooler, where it is evaporated, cooling the chilled water.

Centrifugal compressor chillers using R-134A or R-22 are defined as

positive-pressure machines, while those using R-123 are negative-pressure

machines, based on the evaporator pressure condition. At standard ARI rating

conditions and using R-134A, the evaporator pressure is 36.6 psig and the

Chapter 112

condenser pressure is 118.3 psig, yielding a total pressure increase or lift provided

by the compressor of 81.7 psig. However, for R-123, these pressure conditions

are 25:81 psig in the evaporator and 6.10 psig in the condenser, yielding a total

lift of 11.91 psig.

Mass flow rates for both refrigerants are essentially the same at

approximately 3 lb/min ton. However, due to the significantly higher density of

R-134A, its volumetric flow rate (cfm/ton), which defines impeller size, is over

five times smaller than R-123 volumetric flow rate.

Compressors using R-123 typically use large diameter impellers

(approximately 40 in. diameter) and direct-coupled motors that (at 60Hz) turn

at 3600 rpm. Compressors using R-134A typically use much smaller impellers

(about 5 in. diameter) that are coupled to the motor through a gearbox or speed

increaser and can operate at speeds approaching 30,000 rpm.

The large wheel diameters required by R-123 puts a design constraint on

the compressor and, to reduce the diameter, they typically utilize two or three

impellers in series or stages to produce an equivalent pressure increase.

FIGURE 1.5. Cutaway of typical centrifugal water chiller.


In practice, the flow paths from the outlet of one stage to the inlet of the next

introduce pressure losses that reduce efficiency to some degree.

Since the evaporator in positive-pressure chillers is maintained at a

pressure well above atmospheric, any leaks in the refrigeration system will result

in a loss of refrigerant and the effect of any leaks is quickly evidenced by low

refrigerant levels in the chiller. However, any leaks associated with a negative-

pressure machine result in the introduction of atmospheric air (consisting of

noncondensable gases and water vapor) into the chiller.

FIGURE 1.6. Rotary screw compressor operation. (Courtesy of the American

Society of Heating, Refrigerating, and Air-Conditioning Engineer’s, Inc.,

Atlanta, Georgia.)

Chapter 114

Noncondensable gases create two problems:

1. The compressor does work when compressing the noncondensable

gases, but they offer no refrigerating effect.

2. Noncondensable gases can “blanket” evaporator and condenser tubes,

lowering heat exchanger effectiveness.

Noncondensable gases can lower the efficiency of the chiller by as much as 14%

at full load.

Moisture introduced with atmospheric air is a contaminant that can allow

the formation of acids within the chiller that can cause serious damage to motor

windings (of hermetic motors) and bearings.

To remove potential noncondensable gases and moisture from negative-

pressure chillers, these chillers are furnished with purge units. While purge units

are very efficient at separating and venting noncondensable gases and moisture

from the refrigerant, they are not 100% efficient and some refrigerant is vented to

the atmosphere each time the purge unit operates. Additionally, to reduce the

potential for leaks when chillers are off, the evaporator should be provided with

an external heater to raise the refrigerant pressure to above atmospheric.

The energy requirement for a rotary compressor chiller at peak load is a

function of (1) the required leaving chilled water temperature, and (2) the

temperature of the available condenser water. As the leaving chilled water

temperature is reduced, the energy requirement to the compressor increases, as

summarized in Table 1.4. Similarly, as the condenser water temperature

increases, the compressor requires more energy (see Chap. 10). Thus, the

designer can minimize the cooling energy input by utilizing a rotary compressor

chiller selected to operate with the highest possible leaving chilled water

temperature and the lowest possible condenser water temperature.

1.3.3. Electric-Drive Chillers

Chiller efficiency is typically defined in terms of its coefficient of performance

(COP). The COP is the ratio of output Btus divided by the input Btus. If the

nominal rating of the chiller is 1 ton of refrigeration capacity, equivalent to

12,000Btu/hr output, and the input energy is equivalent to 1 kW, or 3,413Btu/hr,

the resulting COP is 12,000/3,413 or 3.52.

Air-cooled reciprocating compressor water chillers have a peak load power

requirement of 1.0–1.3 kW/ton, dependingoncapacity and ambient air temperature.

Thus, the peak load COP for these units will range from 3.52 to 2.70. Typical rotary

compressor water chillers with water-cooled condensing have a peak load power

requirement of 0.5–0.7 kW/ton, resulting in a COP of 7.0–5.0.

The energy consumption by a rotary compressor chiller decreases as the

imposed cooling load is reduced, as shown in Figure 1.7. These chillers operate

efficiently at between approximately 30 and 90% load and most efficiently


between 40 and 80% load. Under these conditions, the gas flow rate is reduced,

yet the full heat exchange surface of the cooler and condenser are still available,

resulting in higher heat transfer efficiency.

Below about 30% load, the refrigerant gas flow rate is reduced to the point

where (1) heat pickup from the motor and (2) mechanical inefficiencies have

stabilized input energy requirements.

The vast majority of electric-drive rotary compressor water chillers utilize a

single compressor. However, if the imposed cooling load profile indicates there will

TABLE 1.4. Rotary Chiller Input Power as a Function of

Chilled Water Supply Temperature

Leaving chilled water temperature

(8F)Compressor input power

(approx. % change)

41 þ7

42 þ5

43 þ2

44 0

45 22

46 26

47 28

48 212

49 216

FIGURE 1.7. Typical rotary compressor part load performance.

Chapter 116

be significant chiller usage at or below 30% of peak load, it may be advantageous to

use a dual compressor chiller or multiple single compressor chillers.

The dual compressor chiller typically uses two compressors, each sized

for 50% of the peak load. At 50–100% of design load, both compressors

operate. However, if the imposed load drops below 50% of the design value,

one compressor is stopped and the remaining compressor is used to satisfy the

imposed load. This configuration has the advantage of reducing the inefficient

operating point to 15% of full load (50% of 30%), reducing significantly the

operating energy penalties that would result from a single compressor

operation.

Negative-pressure chillers are typically somewhat more efficient than

positive-pressure chillers. A peak load rating of 0.5 kW/ton or less is available for

negative-pressure chillers, while positive-pressure chiller ratings below

0.55 kW/ton are difficult to obtain.

Positive-pressure chillers tend to be smaller and lighter than negative-

pressure chillers, which can result in smaller chiller rooms and lighter structures.

Negative-pressure chillers generally have a higher first cost than positive-

pressure machines.

When purchasing a chiller, owners must decide if the improved efficiencies

of negative pressure chillers are worth the additional first cost, the environmental

impact of releasing refrigerants, the higher toxicity of R-123, and the impact of

the phaseout of HCFC refrigerants.

1.3.4. Engine-Drive Chillers

Natural gas and propane fueled spark ignition engines have been applied to rotary

compressor systems. The full-load cooling COP’s for engine-driven chillers are

approximately 1.0 for reciprocating compressors, 1.3–1.9 for screw compressors,

and 1.9 for centrifugal compressors. These low COP’s can be improved if the

engine water jacket heat and exhaust heat can be recovered to heat service hot

water or for other uses.

Engine-drive chillers have been around for many years, but their

application, most typically utilizing natural gas for fuel, has been limited by a

number of factors:

1. Higher first cost.

2. Air quality regulations.

3. Much higher maintenance requirements.

4. Short engine life.

5. Noise.

6. Larger physical size.

7. Lack of integration between engine and refrigeration subsystems.


Since the mid-1980s, manufacturers have worked very hard to reduce these

negatives with more compact designs, emissions control systems, noise

abatement measures, basic engine improvements, and development of overall

systems controls using microprocessors.

However, the maintenance requirements for engine-drive chillers remains

high, adding about $0.02/ton hr to the chiller operating cost. Currently, the

engines used for chillers are either spark-ignition engines based on automotive

blocks, heads, and moving components (below about 400 ton capacity) or spark-

ignition engines using diesel blocks and moving components (for larger

chillers). While the automotive-derivative engines are advertised to have a

20,000 hr useful life, the real life may be much shorter, requiring an engine

replacement every 2 years or so. The diesel-derivative engines require an

overhaul every 10–12,000 hr (equivalent to a diesel truck traveling

500,000miles at 50mph).

Newer engines use lean burn technology to improve combustion and

reduce CO and NOX emissions. By adding catalytic converters to the exhaust and

additional emissions controls, natural gas fired engine drive chillers can meet

stringent California air quality regulations.

Gas engine-drive chillers remain more expensive than electric-drive units

and they have higher overall operating costs, including maintenance costs, (see

Table 1.5). However, engine-drive chillers may be used during peak cooling load

periods to reduce seasonal peak electrical demand charges (see Sec. 2.3).

TABLE 1.5. Chiller Efficiency and Estimated Energy Cost

Electrical input

(kW/ton)

Heat input

(Mbh/ton)

Costa

($/ton-hr) COP Application/compressor type

— 18.5 0.114 0.67 Single stage absorption

— 14.1 0.087 0.85 Two stage absorption

— 12.0 0.074 1.0 Direct-fired absorption

12.0 0.074 1.0 Engine-drive reciprocating

7.5 0.046 1.6 Engine-drive screw

6.3 0.039 1.9 Engine-drive centrifugal

1.2 — 0.086 2.9 Air-cooled pos disp

1.0 — 0.072 3.5 Air-cooled rotary

0.8 — 0.058 4.4 Water-cooled rotary




0.4 0.029 8.8 Water-cooled rotary

a Based on year 2000 U.S. average commercial energy costs: Electricity, $0.072/kWh; Natural Gas,

$0.616/therm (100,000Btu).

Chapter 118

1.3.5. Condensing Medium

The heat collected by the water chiller, along with the excess compressor heat,

must be rejected to a heat sink. Directly or indirectly, ambient atmospheric air is

typically used as this heat sink.

For air-cooled chillers, the condenser consists of a refrigerant-to-air coil

and one or more fans to circulate outdoor air over the coil. The performance of

the condenser is dependent on the airflow rate and the air’s dry bulb temperature.

Air-cooled condenser air flow rates range from 600 to 1200 cfm/ton with a

10–308F approach between the ambient dry bulb temperature and the refrigerant

condensing temperature. For R-22 in a typical HVAC application, the condensing

temperature is about 1058F. Thus, the ambient air temperature must be no greater

than 958F. As the ambient air temperature increases, the condensing temperature

increases and net cooling capacity decreases by about 2% for each 58F increase in

condensing temperature.

Water-cooled chillers typically use a cooling tower to reject condenser heat

to the atmosphere and Chaps. 9–17 of this text address this topic in detail. At the

chiller, with 858F condenser water supplied from the cooling tower, condensing

temperatures are reduced to 94–988F, reducing the lift required of the

compressor and significantly improving the chiller COP compared to air-cooled

machines.

Table 1.5 illustrates the relative efficiency and operating cost for the

various types of electric-drive chillers with both air- and water-cooled

condensing.

1.4. ABSORPTION CHILLERS

Approximately 92% of refrigeration systems utilized for HVAC applications in

the United States are electric-drive vapor compression cycle systems. However,

in some areas, principally in large cities and at some universities and hospital

complexes, large steam distribution systems are available. In years past, this

steam was often cheaper than electricity and was used to provide cooling,

utilizing the absorption refrigeration cycle. This is generally no longer the case

and little or no new absorption cooling is utilized except where a waste heat

source is available, such as with cogeneration or some industrial processes, or

where the use of absorption cooling during peak cooling load periods may allow a

reduction in seasonal electric demand charges (see Sec. 2.3).

The absorption refrigeration cycle is relatively old technology. The concept

dates from the late 1700s and the first absorption refrigeration machine was built

in the 1850s. However, by World War I, the technology and use of reciprocating

compressors had advanced to the point where interest in and development of

absorption cooling essentially stagnated until the 1950s. During this period, the


two-stage absorption refrigeration machine was developed in the United States,

while the direct-fired concept was perfected in Japan and other Pacific-rim

countries.

The fundamental “single stage” absorption cycle is represented in Figure 1.8.

The evaporator consists of a heat exchanger, held at low pressure, with a separate

refrigerant (typically, water) pump. The pump sprays the refrigerant over the tubes

containing the chilled water, absorbs heat from the water, and evaporates as a low-

pressure gas. The low-pressure gas flows to the absorber due to the pressure

differential. The absorber is at a lower pressure than the evaporator is because the

concentrated absorbent solution (typically lithium bromide) exerts a molecular

attraction for the refrigerant. The absorbent solution is sprayed into contact with the

refrigerant vapor. Condensing of the refrigerant occurs because the heat is absorbed

by absorbent. The absorbent, then, is cooled by condenser water.

The absorbent now consists of a dilute solution, due to its having absorbed

water vapor refrigerant. The dilute solution is pumped to the concentrator, where

heat is applied to re-evaporate the refrigerant. The concentrated solution of

absorbent is then returned to the absorber. The refrigerant vapor goes to the

condenser, where it is condensed by the condenser water. To improve efficiency,

a heat exchanger is used to preheat the dilute solution, with the heat contained in

the concentrated solution of the absorbent.

Leaks allow air to enter the refrigerant system, introducing noncondensable

gases. These gases must be removed, or purged, to prevent pressure in the

absorber increasing to the point where refrigerant flow from the evaporator will

stop. The solution in the bottom of the absorber is relatively quiet and these gases

tend to collect at this point. They can be removed through the use of a vacuum

pump, typically called a purge pump,

Absorption chillers are defined as indirect-fired or direct-fired and may be

single-stage or two-stage (and research on a three-stage chiller is currently

underway), as follows:

1. The indirect-fired single-stage machine uses low- to medium-pressure

steam (5–40 psig) to provide the heat for the absorption process. This

type of chiller requires approximately 18,500Btu/hr per ton of cooling

effect, resulting in a chiller COP of about 0.67.

2. The indirect-fired two-stage chiller utilizes high-pressure steam (at

least 100 psig) or high temperature hot water (4008F or higher) and

requires approximately 12,000Btu/hr per ton of cooling effect,

resulting in a chiller COP of 1.0.

3. The direct-fired chiller, as its name implies, does not use steam but

utilizes a natural gas and/or fuel oil burner system to provide heat.

These chillers are two-stage machines with a resulting in an overall

COP of 1.0–1.1.

Chapter 120

FIG

URE1.8.

Single

stagesteam

absorptionchillerschem

atic.


For the indirect-fired units, the overall COP must be reduced to account for the

losses in the steam production in the boilers. With a typical boiler firing efficiency

of 80–85%, this reduces the overall COP for the single stage system to

approximately 0.54 and to approximately 0.80 for the two-stage system.

Because absorption cooling has a COP of only 0.54–1.1, it competes

poorly with electric-drive rotary compressor chillers, as shown in Table 1.5.

Other factors that must be considered for absorption chillers include the

following:

1. Absorption chillers require approximately 50%more floor area than the

equivalent electric-drive (vapor compression cycle) chiller. Addition-

ally, due to their height, mechanical equipment rooms must be 6–10 ft

taller than rooms housing electric-drive chillers. Finally, because the

liquid solution is contained in long, shallow trays within an absorption

chiller, the floor must be as close to absolutely level as possible.

2. Absorption chillers will weight at least twice as much the equivalent

electric-drive chiller.

3. Due to their size, absorption chillers are sometimes shipped in several

sections, requiring field welding for final assembly.

4. While most electric chillers are shipped from the factory with their

refrigerant charge installed, the refrigerant and absorbent (including

additives) must be field installed in absorption chillers.

5. While noise and vibration are real concerns for electric-drive chillers

(see Sec. 6.1), absorption chillers (unless direct-fired) are quiet and

essentially vibration-free.

6. Due to the potential for crystallization of the lithium bromide in the

chiller if it becomes too cool, the condenser water temperature must be

kept above 75–808F.7. An emergency power source may be required if lengthy power outages

are common. Without power and heat input, the chiller begins to cool

and the lithium bromide solution may crystallize. However, because an

absorption chiller has a very small electrical load requirement (usually

less than 10 kW), a dedicated back-up generator is not a major element.

8. The heat rejection rate from the condenser is 20–50% greater than for

the equivalent electric-drive chiller, requiring higher condenser water

flow rates and larger cooling towers and condenser water pumps.

9. Finally, an indirect-fired absorption chiller will be at least 50% more

expensive to purchase than the equivalent electric-drive chiller. Direct-

fired absorption chillers will cost almost twice as much as an electric

machine, and have the added costs associated with providing

combustion air and venting (stack).

Chapter 122

Direct-fired absorption cycle chillers should be carefully evaluated anytime an

engine-drive vapor compression cycle chiller is being considered. Even though

the energy cost for the absorption chiller may be higher, the increased

maintenance costs associated with engine-drive systems may make the

absorption chiller more cost effective.


2

Chiller Configurations

2.1. THE SINGLE CHILLER SYSTEM

The basic chilled water piping configuration for a single chiller is shown in

Figure 2.1. Here a single chiller provides chilled water to the cooling coils

utilizing a single chilled water pump.

For small systems, this configuration has the advantage of lower initial cost,

but does have some basic disadvantages:

1. With the single compressor system, failure of any component

(compressor, pump, or condenser) will result in no cooling being

available. For most facilities this is unacceptable and the use of

multiple chillers allows at least some cooling (50% or more) to be

provided even if one chiller fails. In cases where cooling is critical to

the facility (computer centers, hospital, laboratories, pharmaceutical or

textile manufacturing, etc.), multiple chillers with at least one

redundant chiller are often used. In this case, even if one chiller fails,

100% of the design cooling load can still be met.

2. As discussed in Chapter 1, once the cooling load imposed on a rotary

compressor chiller falls to below about 30% of the chiller capacity, the

efficiency of the chiller begins to decline. Thus, multiple chillers may

allow a better overall capacity-to-load ratio and improved operating

efficiency.

24

2.2. MULTICHILLER SYSTEMS

When it is desirable to have multiple chillers, there are two basic configurations

that can be utilized, series or parallel.

In the series configuration with two chillers, as shown in Figure 2.2, each

chiller is selected to produce half of the required cooling at the full system flow

rate. Thus, half of the total design range is produced by each chiller. Load ratios

other than 50/50 are possible, but this is by far the most common condition

because of control problems with chillers at very small temperature differences.

Table 2.1 summarizes the temperatures at various load conditions for the

configuration shown in Figure 2.2.

FIGURE 2.1. Constant flow, single chiller configuration.

TABLE 2.1. Series Configuration Temperatures

Temp Full load 75% load 50% load 25% Load

TS 44 44 44 44

TR 56 53 50 47

T1 50 50 50 47

Chiller Configurations 25

Series chiller systems are rarely done any longer because this configuration

requires a constant chilled water flow rate at all times, resulting in high pumping

costs. But, if a relatively large temperature difference is required or if there is a

very steady base cooling load, the series configuration may offer some

advantages.

The parallel chiller configuration is far more common. In a two-chiller

configuration, each chiller is typically selected to operate with the same design

range, but with only a half of the total system flow requirement. This again results

in a 50/50 load split, but other load ratios may be selected if dictated by

operational requirements.

2.2.1. One Pump Parallel Configuration

In the one pump parallel chiller configuration shown in Figure 2.3, the overall

system performance and temperature conditions are summarized in Table 2.2.

With this configuration, there is an inherent problem. If both machines

were operated for the full load range (15–100% of peak capacity), by the time the

total system load drops to 30% of full load, each individual chiller would be

operating very inefficiently. Thus, most designers utilize controls to shut off one

chiller when the total system load, as evidenced by the return water temperature,

falls below 50% of full load.

However, with this piping arrangement, if one chiller is not in operation,

chilled water from the operating chiller will mix (blend) with the return water

passing through the nonoperating chiller, effectively raising the chilled water

supply temperature to the system. In many cases, this may not be a problem. But,

generally the interior spaces of large buildings still require more-or-less full

cooling even when the perimeter spaces require no cooling at all. An elevated

chilled water supply temperature may not satisfy these interior load conditions.

To attempt to eliminate the blended supply water problem with the one

pump configuration, some designers have used chiller flow isolation valves, as

FIGURE 2.2. Series chiller configuration.

Chapter 226

shown in Figure 2.4. With this configuration, flow through the nonoperating

chiller is closed off when the chiller is not in operation.

This arrangement results in increased flow through the operating chiller,

but does reduce the blending problem, as illustrated by Table 2.3.

2.2.2. Multiple Pump Parallel Configuration

To ensure that the blended water condition does not occur, the multiple pump

parallel chiller configuration shown in Figure 2.5 is widely used.

With this configuration, each chiller has an individual chilled water pump.

Thus, when one chiller is not operating—one pump is off, flow through the

nonoperating chiller is zero, and no blending results.

Table 2.4 summarizes the performance and temperature conditions for this

configuration at various load conditions.

FIGURE 2.3. One pump parallel chiller configuration.

TABLE 2.2. One Pump Parallel Configuration Flows and Temperatures

Chiller no. 1 Chiller no. 2

%Load Tr Flow (%) T1 Flow (%) T2 Ts

100 56 50 44 50 44 44

75 53 50 44 50 44 44

50 50/56 50 44 50 56a 50

25 50 50 44 50 50a 47

a Chiller no. 2 off.


2.2.3. Primary–Secondary Parallel Configuration

Each of the configurations discussed earlier are essentially constant flow systems

that utilize three-way control valves at the cooling coils. Constant flow systems

circulate the same amount of chilled water no matter what the imposed cooling

load and, consequently, result in high pumping energy costs.

To reduce these costs, the primary–secondary variable flow piping

arrangement illustrated in Figure 2.6 is very common. Here, the production loop

through the two chillers is hydraulically isolated from the distribution loop by a

piping bridge. The bridge is a short section of piping shared by the two loops and

designed to have little or no pressure drop. Thus, flow in either loop is not

affected by flow in the other.

On the primary or “production” loop side, the system acts as a multiple

pump parallel chiller installation, as described in Sec. 2.2.2. Flow in this loop

varies in “steps” as the chillers are staged on or off and their respective pumps are

started and stopped.

FIGURE 2.4. One pump parallel chiller configuration with isolation valves.

TABLE 2.3. One Pump Parallel Configuration with Isolation Valves

Flows and Temperatures



100 56 50 44 50 44 44

75 53 50 44 50 44 44

50 50/53 50/67 47 50/0 N/Aa 47

25 47 67 44 0 N/Aa 44

aChiller no. 2 off.

Chapter 228

However, in the secondary or distribution loop, the cooling coils utilize two-

way control valves and the pump(s) is fitted with a variable speed controller that

modulates the flow from0 to 100%as a function of the imposed cooling load. Thus,

this loop has variable flow, but a constant range. At any load condition, the supply

water temperature is the same as the water temperature leaving the chiller(s), as

long as the production loop flow rate equals or exceeds the distribution loop flow

rate.

Table 2.5 summarizes the performance and temperature conditions for this

configuration at various load conditions.

2.2.4. Variable Primary Flow Parallel Configuration

Since the early 1990s, rotary compressor water chillers have been manufactured

with integral digital control systems. This represents a significant improvement in

chiller control and allows the use of the variable primary flow piping arrangement

FIGURE 2.5. Multiple pump parallel chiller configuration.

TABLE 2.4. Multiple Pump Parallel Configuration Flows and

Temperatures



100 56 50 44 50 44 44

75 53 50 44 50 44 44

50 50 50 44 50/0 N/Aa 44

25 47 50 44 0 N/Aa 44

aChiller no. 2 off.


shown in Figure 2.7 to save even more in pumping energy (typically,

approximately 15–20%).

The flow through the chillers is varied in response to the system load.

A bypass valve opens to maintain minimum flow equivalent to about 30% of the

flow through the largest chiller. This system has certain operational limitations

since rapid load fluctuations and flow changes may result in fluctuation in the

chilled water supply temperature, but is a viable option where critical chilled

water temperature control is not required.

Basic criteria for this system include the following:

1. Flow through each chiller must be maintained in the range 3–11 fps.

Consequently, good-quality flow meters, which must have periodic

recalibration, are required for each chiller.

2. If flow rates through the chiller change too rapidly, the chiller controls

cannot keep up. Therefore load flucuations must be limited to not more

than 30% per minute.

One misconception about the variable primary flow configurations is that the

chillers in this type of system will operate more efficiently. With variable flow

through the evaporator, the LMTD of the evaporator remains constant but the

convective heat transfer coefficient decreases with flow, thus reducing the heat

transfer effectiveness. With constant flow through the evaporator, the LMTD

falls as the entering water temperature falls, but the convective heat transfer

coefficient remains constant since flow remains constant. Thus, power

consumption by a chiller will be essentially the same whether evaporator

water flow is constant or variable.

FIGURE 2.6. Primary–secondary parallel chiller configuration.

Chapter 230

TABLE2.5.

Primary–SecondaryConfigurationFlowsandTem

peratures

Chillerno.1

Chillerno.2

%Load

Tr

%Schw

%Pchw

Tm

Flow

(%)

T1

Flow

(%)

T2

Ts

100

56

100

100

56

50

44

50

44

44

75

56

75

100

53

50

44

50

44

44

50

56

50

50

56

50

44

50/0

N/A

a44

25

56

25

50

53

50

44

0N/A

a44

aChillerno.2off.


Another misconception about these systems is cost. Often the literature

will state that variable primary flow systems are less costly than primary–

secondary systems because one set of pumps is eliminated. However, the cost

of high-quality flow meters (and the additional piping to ensure they are

installed properly for accurate reading), the bypass flow control valve, and the

more complex system controls will generally offset pump cost savings.

For large systems with high flow rates and long distribution piping runs, the

additional savings in pumping energy produced by this configuration may be

advantageous. But, it is amuchmore complex system and has operating limitations

previously discussed. So, for most systems, the simpler primary–secondary system

configuration remains the recommended approach for the majority of multiple

chiller systems, though variable primary flow should be evaluated for larger

systems with long distribution runs and more sophisticated operators.

2.3. SYSTEM PEAK COOLING LOAD AND LOADPROFILE

One primary goal for chiller operation is to produce chilledwater as economically as

possible. To meet this goal, not only must individual chillers be efficient, but the

entire multichiller system must be designed to function efficiently.

FIGURE 2.7. Variable flow primary parallel chiller configuration.

Chapter 232

The first step in evaluating the requirements for a chilled water system is to

determine the peak cooling load requirement. However, the peak loadwill occur only

for a very limited number of hours during the year. On an annual basis, the imposed

load will vary based on the time of day due to occupancy patterns, solar heat gains,

and diurnal temperature swings, etc. and the time of year due to solar and temperature

seasonal variations. Another factor that influences the load imposed on the chilled

waterplant is theamountof “core” building area thatwill require coolingyear-around

andwhether an airside economizer cycle is installed to provide “free” cooling during

the winter. With all of these variables, every chilled water system will operate far

more hours at a part-load condition than it will operate at the peak load condition.

Thus, in addition to determine the peak load to be imposed on the system, it

is also necessary to determine the part-load profile, either as a function of time,

outdoor temperature, or both.

The best way to illustrate the process of estimating a system cooling load

profile is by example:

1. Assume a chilled water system serves a building with the peak load

characteristics summarized in Table 2.6. This peak load is based on the

building being fully utilized at 4:00 PM on an August afternoon.

2. The next step is to evaluate the peak day load profile by determining the

loads imposed eachhourof the designday as shown inTable 2.7.Note that

there is both an “occupancy variable” for lighting and other internal heat

TABLE 2.6. Example of Peak Heat Gain Components

(4:00 PM, August 31, 948F DB/788F WB Outdoor

Conditions)

Component Heat gain (Btu/hr)

Opaque walls and roof conduction 165,000

Window conduction 210,000

Window solar radiation 490,000

Internal

People, sensible 25,000

People, latent 20,000

Lights 900,000

Equipment/appliances 75,000

Ventilation

Sensible 427,000

Latent 390,000

Total 2,702,000

225 tons


TABLE2.7.

Exam

ple

Estim

ated

DesignDay

Load

Profile

Heatgain(1000Btu/hr)

Hour

O.A.temp

Walls/roof

Window

cond.

Window

solar

Internal

Vent.

Total

Totaltons

178

26

33

0102

129

29

24

276

911

0102

43

165

14

374

29

211

0102

243

39

3

476

911

0102

43

165

14

577

17

22

0102

86

227

19

678

26

33

0102

129

290

24

780

43

55

10

102

215

425

35

881

52

66

50

102

258

528

44

982

61

77

60

750

301

1249

104

10

84

78

99

80

900

387

1544

129

11

86

96

122

100

1020

473

1811

151

12

88

113

144

150

1020

559

1986

166

13

90

130

166

200

1020

645

2031

169

14

92

148

188

300

1020

731

2387

199

15

93

156

199

400

1020

774

2549

212

16

94

165

210

490

1020

817

2702

225

17

93

156

199

480

1020

774

2549

212

18

92

148

188

400

900

731

2367

197

19

90

130

166

350

900

645

2191

183

20

86

96

122

150

700

473

1541

128

21

84

78

99

90

500

387

1154

96

22

82

61

77

0102

301

541

45

23

81

52

66

0102

258

470

39

24

80

43

55

0102

215

415

35

Chapter 234

gains, a “solar variable” due to the day/night cycle, and a “temperature

variable” due to the diurnal temperature profile over the 24-hr period.

3. Next, examine the load imposed with winter outdoor conditions. While

the occupancy variables will be the same as for the summer month, the

solar and temperature variables will be markedly different. To determine

the imposed chilled water load, the building may be considered to have a

perimeter zone, in which heat losses through walls, roofs, and windows

offset internal heat gains, and a core zone, in which only roof loads can be

considered to offset internal heat gains. Obviously, the cold ventilation air

can help offset any imposed space cooling loads.

In Table 2.8, the building is assumed to have 50/50 core/perimeter

area relationship and the loads are evaluated for different outdoor

temperatures for both “occupied” and “unoccupied” conditions.

4. Finally, consider the impact of airside economizer cycles. If all of the

air-handling units are equipped with economizers, there is no

requirement for chilled water when the outdoor temperature is below

the cooling coil discharge temperature, typically 50–558F. If airsideeconomizers are not provided, then either chilled water must be

provided by running the chillers and/or using a waterside economizer

cycle (see Sec. 14.2).

From this evaluation, both the peak load and the minimum load to be imposed on

the chilled water system are determined.

1. In this example, the peak load is 225 tons.

2. If the example building were an office building, in all probability the air

systems would be shut down when the cleaning crew is finished at

about 8:00 PM (hour 20) and not started until about 6:00 AM (hour 6).

However, for a lab building, the air systems will operate 24 hr per day.

Therefore, from Table 2.7, the minimum load imposed on the chilled

water system would be as low as 3 tons if the building were cooled 24 hr

per day, but 24 tons if the building is shut down at night.

3. Based on Table 2.8, the buildingwill require cooling by the chilledwater

system during the occupied period down to 408F outdoor air

temperature. If airside economizers are in use, the minimum winter

occupied cooling load is 76 tons, occurring at 608F outdoor temperature.

2.4. SELECTING WATER CHILLERS2.4.1. Basic Chiller Requirements

The peak cooling capacity and corresponding input energy for water chillers is

rated in accordance with the Air-conditioning and Refrigeration Institute (ARI)


TABLE2.8.

Exam

ple

WinterCoolingRequirem

ents

Perim

eter

zones

heatgain/loss

(Btu/hr)

O.A.

temp

bin

Occupancy

Walls/Roof

Window

cond.

Window

solar

Internal

Core

zones

internal

heatgain

(Btu/hr)

Ventilationheat

gain/loss

(Btu/hr)

Totalheat

gain/loss

(Btu/hr)

Total

tons

0Occupied

2618

2788

490

510

510

21601

21497

Unoccupied

2618

2788

051

51

21601

22956

10

Occupied

2536

2673

490

510

510

21388

21097

Unoccupied

2536

2673

051

51

21388

22495

20

Occupied

2454

2578

490

510

510

21174

2696

Unoccupied

2454

2578

051

51

21174

22104

30

Occupied

2371

2473

490

510

510

2961

2295

Unoccupied

2371

2473

051

51

2961

21703

40

Occupied

2289

2368

490

510

510

2747

106

9

Unoccupied

2289

2368

051

51

2747

21302

50

Occupied

2206

2263

490

510

510

2534

507

42

Unoccupied

2206

2263

051

51

2534

2901

60

Occupied

2124

2158

490

510

510

2320

908

76

Unoccupied

2124

2158

051

51

2320

2531

Chapter 236

Standard 550/590-98, Water-Chilling Packages Using the Vapor Compression

Cycle. Chillers must be constructed and installed in accordance with ASHRAE

Standard 15, Safety Code for Mechanical Refrigeration. Since the evaporator and

condenser for R-22 or R-134A chillers operate at pressures exceeding 15 psig,

they are defined as pressure vessels and must be constructed in accordance with

Section 8 of the American Society of Mechanical Engineers (ASME) Boiler and

Pressure Vessel Code. And, all chillers must comply with the requirements of

Underwriters Laboratory (UL) Standard 465.

The rated capacity of a water chiller, determined in accordance with ARI

550/590-98 is based on a standard set of operating conditions, as follows:

If any of these conditions vary for a specific project or location, the

manufacturers will provide a specific cooling rating for the required conditions.

Typically, if the chilled water temperature increases, chiller capacity increases. If

the entering condenser water or air temperature decreases, chiller capacity

increases. Generally, capacity does not vary with changes in range.

2.4.2. Part-Load Efficiency

As shown in Sec. 2.3, chillers rarely operate for long at maximum capacity. To

provide data to evaluate chiller part load performance, ARI 550/590-98 requires

chiller manufacturers to determine the part-load operating performance

characteristics, as follows:

1. Integrated Part-Load Value (IPLV )—This rating method requires the

manufacturer to determine the input energy requirement for each

chiller at the four operating conditions defined in Table 2.9. The

“weighing factor” defined in Table 2.9 represents the percentage of the

hours of the cooling season at which the chiller is expected to operate at

the defined rating point.

Since chiller energy performance will improve with reduced load

and reduced condensing temperatures, the IPLV can be used to

compare alternative chiller seasonal energy consumption using a

common rating method. All other factors being equal, the chiller with

the lower IPLV will have lower seasonal operating costs.

Chilled water supply temperature 448F

Chilled water temperature range/flow rate 108F/2.4 gpm/ton

Evaporator fouling factor 0.0001

Condenser water supply temperature/flow rate (water-cooled) 858F/3.0 gpm/ton

Condenser fouling factor 0.00025

Entering air temperature (air-cooled) 958F


The weighing factors used in the ARI standard are based on (1)

weighted average weather data for 29 U.S. cities, and (2) weighted

hours of use at different building operating scenarios.

2. Nonstandard Part-Load Value (NPLV )—The IPLV rating is only

directly applicable to the single chiller installation where the condenser

water supply temperature is allowed to fall as the outdoor wet bulb

temperature falls.

Anytime multiple chillers are used, particularly if careful control

allows the chillers to be used at or near their optimum efficiency range

40–80% of rated capacity, the IPLV has little validity. Likewise, if

control of the condenser water temperature control method does not

allow the supply temperature to go as low as 658F, then the IPLV is not

valid. And, in fact, at least two chiller manufacturers recommend

limiting the supply condenser water temperature to 658F and 758F,respectively.

Therefore, to compare the seasonal energy performance of

alternative chillers, it is necessary to define the specific weighing

factors and condensing conditions to be applied to compute an NPLV.

(See Chap. 8 for further discussion of this topic.)

Energy codes in the United States are, by Federal law, required to be equivalent to

ASHRAE/IESNA Standard 90.1, Energy Standard for Buildings Except Low-

Rise Residential Buildings. The 1999 version of this standard requires that water

chillers, when selected for rated conditions, have minimum peak load COP’s and

rated IPLVs shown in Table 2.10. (Typically, though, it is more cost effective to

purchase chillers that are more efficient than these minimum values.)

2.4.3. Load vs. Capacity

One of the major chiller manufacturers estimates that, for the typical chiller

installation, 25% of the total owning and operating cost over the life of the chiller

is related to the cost of designing, purchasing, and installing the chiller, while

TABLE 2.9. IPLV Rating Conditions (ARI Standard 550/590-98)

Rating

point

Percent of

full load

Weighing

factor (%)

Water-cooled condenser

water temperature (8F)

Air-cooled entering

air temperature (8F)

A 100 1 85 95

B 75 42 75 80

C 50 45 65 65

D 25 12 65 55

Chapter 238

75% of the total cost is consumed by energy expenses and maintenance costs.

Therefore, while the chiller must be selected to meet the peak cooling load, or its

assigned part of the peak load, it is far more important to select each chiller to

operate as efficiently as possible over the full range of part load conditions that

can be expected.

If the chiller “plant” is to consist of a single chiller, it is only necessary

to select the chiller to meet the peak cooling load and then select the most

efficient chiller (the chiller with the lowest IPLV or NPLV) within the scope

of the available capital funds. But, when multiple chillers are to be selected,

the selection of each chiller becomes more complex. But, the basic

requirement is, as closely as possible, to match capacity to load across the full

load profile.

As discussed in Sec. 2.1, it is usually desirable to have multiple chillers

available to satisfy the cooling load imposed on a chilled water system. There are

several aspects that must be considered when selecting the chillers for a

multichiller system:

1. Compressor Type—For peak cooling loads below 100–125 tons, rotary

compressor chillers are expensive and only a single chiller could be used

since this is the minimum capacity available for this type of compressor.

Therefore, the only choice is to use reciprocating compressors.

For chillers less than about 20 tons capacity, the only way to have a

multichiller system is to use multiple, modular water chillers that are

available in the 2–10-tons capacity range. For larger capacities, multiple

reciprocating compressors are utilized for a single chiller, with asmany as

four compressors used at 100–125-tons capacity. In effect, since each

TABLE 2.10. Minimum Water Chiller Efficiency Requirements from ASHRAE/IESNA

Standard 90.1-1999

Chiller type

Capacity

(tons)

Peak load

COP

IPLV

(kW/ton)

Reciprocating, air-cooled All 2.80 1.256

Reciprocating, water-cooled All 4.20 0.756

Screw, water-cooled ,150 4.45 0.781

151–299 4.90 0.710

$300 5.50 0.628

Centrifugal, water-cooled ,150 5.00 0.703

151–299 5.55 0.634

$300 6.10 0.576

Absorption, 2-stage, direct or indirect,

water-cooled

All 1.00 —


compressor can be arranged with an independent refrigerant circuit, this

arrangement yields the same reliability and redundancy as a multichiller

system and is perfectly acceptable.

Between 100 and 200 tons peak cooling load, two or more

reciprocating compressor chillers can be used. Above 200 tons, rotary

compressor systems begin to become cost effective.

2. Condensing Medium—For peak cooling loads below 100–125 tons, the

capital cost and increased maintenance requirements for a water-cooled

system are rarely justified and the chiller(s) will be air-cooled. Above

200 tons peak cooling load, and with the use of rotary compressor

chillers, the addition of cooling towers to the system to provide for

water-cooled condensing become justifiable. Between 100 and 200 tons

peak cooling load, it becomes a matter of the owner’s ability to deal

with the maintenance requirements of a cooling tower system, owner

preference (if any), and the capital funds available.

3. Symmetrical Chiller Capacity—With this approach, all of the chillers

are sized for equal capacity. The number of chillers and, thus, the size

of the chiller “module” is based on the minimum anticipated load. Once

the plant load is reduced to below the capacity of a single chiller, we

want that chiller to operate in an efficient region (i.e., above 30%

capacity) as long as possible.

For example, a building with a peak cooling load of 750 tons and a

minimum cooling load of 100 tons could be served by three chillers,

each rated at 250 tons. At 100 tons, a single chiller would operate at

40% capacity, which is still within the efficient region.

Had two 375 ton chillers been selected, the last chiller on line

would operate at 27% of its capacity to meet the minimum cooling load

requirement.

In this example, if the minimum cooling load had been 20 tons,

even the 250 ton chiller would not have been able to fully respond.

Since chiller minimum capacity is typically about 15% of the peak

capacity, the chiller would have cycle off under its internal controls

when the load fell below about 40 tons.

4. Asymmetrical Chiller Capacity—There is no engineering rule that says

that all chillers in a multichiller system have to be the same size. While

there may be some maintenance advantages (common parts, etc.),

different sized chillers can be operated together.

In the previous example, the 750 peak load could have been met

with a 600- and a 150-ton chiller. Both chillers would operate to

produce 750 tons at peak load conditions, but the smaller chiller would

meet the minimum load operating at 50% of its peak capacity, which

would normally be a very efficient operating point.

Chapter 240

Table 2.11 summarizes the load vs. capacity relations for the

symmetrical and asymmetrical examples. The asymmetrical arrange-

ment results in potentially lower operating cost and the ability to satisfy

a lower minimum cooling load. If a single chiller fails with the

symmetrical design, two-thirds of the peak capacity is still available.

However, if the 600 ton chiller in the asymmetrical arrangement fails,

only 20% of the plant capacity is available.

Another common asymmetrical configuration is the 60/40 split

with two chillers. In this case, one chiller is sized for 40% of the load

while the larger chiller is sized for 60% of the load. In many

applications, this split will allow better capacity-to-load matching and

improved performance.

2.5. MIXED ENERGY SOURCE CHILLER SYSTEMS

For smaller chilled water systems (less than 800–1000 tons), the lower average

operating cost for electric-drive chillers generally eliminates alternative energy

source considerations. However, for larger plants, particularly where electrical

demand charges are high, have the potential of using absorption and/or engine-

drive chillers as part of the multichiller “mix.”

As discussed in Chap. 1, both absorption chillers and engine-drive chillers

are less efficient than electric-drive chillers and, generally, have higher operating

and maintenance costs. However, if the seasonal cooling load profile is such that

TABLE 2.11. Comparison of Symmetrical vs. Asymmetrical Chiller Plant Performance

Symmetrical plant output (tons)

Asymmetrical plant

output (tons)

%Peak load Load (tons) Chiller 1 Chiller 2 Chiller 3 Chiller 1 Chiller 2

100 750 250 250 250 600 150

90 675 225 225 225 540 135

80 600 200 200 200 600 Off

70 525 175 175 175 525 Off

60 450 225 225 Off 450 Off

50 375 188 187 Off 375 Off

40 300 150 150 Off 300 Off

30 225 225 Off Off 225 Off

20 150 150 Off Off Off 150

10 75 75 Off Off Off 75


higher cost cooling methods can be used for a limited number of hours to reduce

the peak electrical demand and associated demand charges, usually imposed on

an annual basis, the net result is an operating cost savings. The first step in

determining if an energy source in addition to electricity should be considered is

to understand electric utility rate schedules and the costs of alternative fuels.

All electric rates divide the allocation of cost into two components, demand

and consumption. Demand, in terms of kW, represents the highest use of

electricity over a specified period time, typically 15–30min, during each

monthly billing period. Consumption, in kWh, is the total use of electricity during

the monthly billing period.

For the electric utility, the demand charge represents the cost incurred by

the utility in having sufficient capacity available to meet each customer’s

maximum need, even though the maximum need may occur only for a limited

number of hours each year. In essence, it represents the utility’s attempt to

equitably distribute the cost of capital recovery for their plants, distribution lines,

substations, etc. to those requiring the availability of this capacity.

For most electric utilities, their peak system demand occurs during the

summer because of air-conditioning use. Thus, demand charges are computed not

only on the basis of the maximum use of electricity in each month, these charges

may be imposed on an annual basis via ratchet clauses in the electric rate.

A typical ratchet clause will define the “billing demand” in each month as

the greater of the actual demand incurred during that month or the maximum

demand incurred in the previous 11months. Other rates may use only an 80%

ratchet, setting the billing demand in each month as the greater of the actual

demand incurred during that month or 80% of the maximum demand incurred in

the previous 11months. Some will separate summer peak demands and winter

peak demands to define the minimum billing demand in each month.

To further address the impact of a customer’s peak demand on the utility’s

system demand, on-peak and off-peak periods are typically specified by the utility

on a daily basis. For example, one utility defines their on-peak summer period to

be from 10 AM to 8 PM, Monday through Friday, during the months of June,

July, August, and September. During the on-peak period, demand charges may

exceed $20.00 per kW, while off-peak demand charges might be only $3.00–

$4.00.

Additionally, utilities impose further peak period cost penalties by

establishing higher consumption charges during on-peak periods relative to

off-peak periods. Power may be $0.14/kWh during the on-peak period, but only

$0.04 during the off-peak period.

Cooling systems, because of their high demand with correspondingly low

energy consumption, have a particularly bad impact on the electric utility’s

overall load factor (the ratio of peak capacity to average capacity required). By

reducing the peak load imposed on the electric utility or by shifting cooling

Chapter 242

energy consumption to off-peak periods, the user is effectively helping the utility

improve its load factor. This benefit can be reflected in the rate schedules and

contract terms that can be negotiated with the utility and can reduce the overall

cost of chilled water production.

The most common approach to reducing electrical energy costs for cooling

is to utilize an alternative energy source during on-peak periods. Alternative

energy sources that can be considered for used in a mixed energy source chiller

system include the following:

1. Natural Gas and/or Fuel Oil—Natural gas can be used in direct-fired

absorption chillers, smaller spark ignition engine-drive chillers, or very

large gas turbine-drive chillers. No. 2 fuel oil can be used in direct-fired

absorption chillers and diesel engine-drive chillers.

Natural gas rate schedules fall into two general categories: “firm”

gas rates and “interruptible” gas rates. Firm gas rates are based on the

utility having gas available at all times, regardless of the load on the gas

distribution system. Interruptible rates are used when the facility allows

the gas utility to cut off their gas supply during peak distribution load

periods. Firm gas rates are typically much higher than interruptible

rates. Since the gas utility has its peak distribution loads during the

winter, gas used during the summer, even if purchased under an

interruptible rate, is a very reliable energy source.

For most gas utilities, the summer distribution load is far less than

their winter distribution load and it is of great economic advantage to

the gas utility if it can add summer load without increasing the winter

peak. Cooling is an ideal candidate for additional summer load and,

therefore, some utilities will offer discounts for purchase of “summer

gas.”

Fuel oil usually is a consideration only if the fuel storage facilities

are already in place to serve the boilers and/or emergency generators.

The costs and environmental considerations associated with large fuel

oil storage facilities are significant and generally eliminate this option

for new installations.

Fuel oil is purchased in bulk lots (usually by the “tanker load” of

approximately 8000 gallons) on the “spot” market. Therefore, the

facility manager must continually shop for the best price.

2. District Steam or High-Temperature Hot Water—Larger campus

facilties such as universities, military bases, and hospitals may have

steam or high-temperature hot water heating distribution systems.

Since these systems are lightly loaded during the summer, either steam

or hot water can be used in absorption chillers to provide chilled water

during on-peak periods.


The cost of steam or high-temperature hot water must include the

cost of the fuel, discounted by the firing inefficiencies and distribution

losses in the system. Summer COP’s for a central boiler plant may be as

low as 0.5–0.7; so careful study is required to determine the real cost of

steam or hot water delivered to the chiller(s).

3. Waste Heat from Industrial Processes—Heat, if sufficiently hot and if

available in the required quantities, can be used to produce steam or

high-temperature hot water for absorption cooling.

As discussed in Chap. 1, no type of chiller is more efficient than the electric-drive

chiller. Additionally, absorption and engine-drive chillers cost more than

equivalent electric-drive chillers to purchase and install. So, the economic

viability for use of an alternative fuel hinges on using an inefficient, high-cost

method of cooling for relatively short periods to save enough on-peak and ratchet

electrical charges associated with an equivalent amount of electric-drive cooling

to pay for the increased capital requirements within an acceptable timeframe.

Typically, this does not happen unless the cost of alternative energy sources

is very low or additional cooling capacity is required anyway and only the

incremental cost increase for alternative fuel systems must be recovered. In the

first case, the energy cost must be validated and its reliability evaluated. In the

second case, the capital cost difference between electric chillers and absorption

or engine-drive chillers must be determined. In either case, careful analysis is

required.

Chapter 244

3

Chilled Water System Elements

3.1. CHILLER PLACEMENT AND INSTALLATION

An air-cooled water chiller must be located outdoors in order to use atmospheric

air as its heat sink. Location and installation of an air-cooled chiller must address

the following considerations:

1. Prevent recirculation of discharge air by locating the chiller with

sufficient clearance to adjacent buildings, shrubbery, and other equipment

to allow unobstructed air movement into and out of the condenser section.

Discharge is obviously warmer than the ambient air and, if allowed to

recirculate and enter the condenser, chiller capacity is reduced.

Screening, either by plantings or screen walls, must be at least

10 ft away from the chiller, be no higher than the top of the chiller, and

have at least 50% open area.

2. Community noise must be addressed. And, for rooftop applications,

vibration must be considered. See Section 6.1 for a detailed discussion

of these topics.

3. Service access must be considered. Ground locations are preferable to

rooftop locations simply because access is easier. (Compressor

replacement for a rooftop chiller can be complex, expensive, and

aggravating due to the need to lift a heavy compressor to the roof level.)

45

Water-cooled chillers are located indoors in a mechanical equipment room or

chiller room. Chiller rooms must be separate from boiler rooms and require space

for the chiller(s), chilled water pump(s), condenser water pump(s), and the

condenser water treatment system (see Chap. 13). ASHRAE Standard 15,

adopted by most building code authorities, classifies a chiller room as a

refrigeration machinery room if the amount of refrigerant in the largest chiller

exceeds certain specified levels. If so classified, the room must be constructed to

higher standards, including specialized ventilation systems, and a detailed code

review for these spaces is critical.

Chillers are big, heavy, and noisy and all of these conditions must be taken

into account when locating and sizing the chiller room:

1. Table 3.1 lists the operating weights for typical water-cooled rotary

compressor chillers. Obviously, it is usually easier to deal with these

concentrated weights in a ground floor or basement location than in the

upper stories of a building.

2. Noise levels as high as 80–90 dBA can be produced by a chiller during

part load operation and the chiller room must be designed to contain the

noise. While vibration by a rotary compressor chiller is very low, it

must also be considered and addressed. Again, see Section 6.1 for a

more detailed discussion of these topics.

3. The chillers, pumps, water treatment equipment, etc. in the chiller room

must be located and arranged to provide for adequate access for

maintenance.

Figure 3.1 illustrates the piping and installation requirements for a typical rotary

compressor chiller installed at or below grade level.

TABLE 3.1. Approximate Chiller Operating Weight

Capacity (tons) Approximate operating weight (lbs) Condensing type

25 3,000 Air-cooled

50 5,000

75 7,500

100 10,000

150 12,000–17,000 Water-cooled

200–300 15,000–20,000

300–500 20,000–28,000

500–700 22,000–30,000

700–900 26,000–44,000

900–1,200 42,000–66,000

1,200–1,500 57,000–71,000

1,500–2,000 68,000–95,000

Chapter 346

FIG

URE3.1.

Typical

chillerpiping.

Chilled Water System Elements 47

3.2. CHILLED WATER PIPING3.2.1. Piping Materials and Insulation Requirements

Chilled water distribution systems are assembled from commercially available

piping materials, most commonly steel and copper.

Steel Pipe—This is the most common above-ground piping type and is

defined by its wall thickness, called schedule, and its finish. Up through

12 in. pipe size, Schedule 40 piping is normally used for chilled (and

condenser) water applications. For piping 14 in. and larger, most

designers standardize on piping with a 0.375 in. wall thickness and

ignore the pipe schedules. All chilled water piping is constructed of

piping with a mill finish and is referred to as “black.” Piping 2 in. and

smaller is usually assembled by threaded joints, while piping 2-1/2 in.

and larger is usually assembled by welding. Table 3.2 lists the physical

properties of steel piping commonly used in water systems.The fittings

for steel pipe with screw connections are typically constructed of cast

iron rated for up to 125 psig working pressure. Wrought steel

“buttwelding” fittings are used for welded piping.

TABLE 3.2. Properties of Commercial Steel Pipe

Pipe

size

(in.)

Outside

diameter

(in.)

Wall

thickness

(in.)

Pipe

schedule

1/2 0.840 0.109 40

3/4 1.050 0.113 40

1 1.315 0.133 40

1-1/4 1.660 0.140 40

1-1/2 1.900 0.145 40

2 2.375 0.154 40

2-1/2 2.875 0.203 40

3 3.500 0.216 40

4 4.500 0.237 40

6 6.625 0.280 40

8 8.625 0.322 40

10 10.750 0.365 40

12 12.750 0.406 40

14 14.000 0.375 30

16 16.000 0.375 30

18 18.000 0.375 30

20 20.000 0.375 20

24 24.000 0.375 20

Chapter 348

Copper Tubing—The cost of copper tubing is higher than that of steel pipe,

but the installation labor cost for smaller sizes, 2 in. and less, is much less

than that of steel. Therefore, most designers allow copper tubing to be

used in water systems for these smaller sizes at the contractor’s

option.To prevent galvanic corrosion that is initiated by dissimilar

metals in direct contact, connection between steel and copper piping

materials must be electrically separated by a dielectric union. This fitting

uses rubber gasketing to separate the two materials.The wall thickness of

copper tubing is indicated by its type, defined as Types K, L, and M in

decreasing order of wall thickness (Table 3.3). Typically, Type K or L

tubing can be used for chilled water piping, but the Type M wall

thickness is too thin.Fittings for copper tubing are wrought copper and

the tubing and fittings are assembled by soldering the joints.

Cast or Ductile Iron—For underground piping, cast iron piping or ductile

iron piping is routinely used. Iron piping is rated for application in terms of

“pressure class” and for most chilled water systems Class 350 is used up

through 12 in. pipe size, Class 250 for 14–20 in. pipe, and Class 150 for

piping 24 in. and larger.Fittings are cast gray or ductile iron and the fittings

and piping are assembled with gasketed mechanical pressure joints.

PVC or CPVC—As an alternative to iron piping for underground chilled

water distribution, PVC or CPVC piping can be used. PVC and CPVC

piping wall thickness is defined in terms of schedules, much like steel

pipe. For chilled water applications, Schedule 40 piping is normally

used.Fittings are also formed of PVC or CPVC and are joined by solvent

soldering up through 4 in. pipe size and by gasketed mechanical pressure

joints for larger sizes.

All above-ground piping must be insulated to reduce unwanted heat gain and to

prevent surface condensation, since chilled water temperatures are normally well

below ambient air dewpoint temperatures during the cooling season. While as little

as 1 in. of insulationmay serve to limit heat gains (and tomeet current energy codes),

TABLE 3.3. Properties of Commercial Copper Tubing

Pipe

size (in.)

Outside

diameter (in.)

Type K wall

thickness (in.)

Type L wall

thickness (in.)

1/2 0.625 0.049 0.40

3/4 0.875 0.065 0.045

1 1.125 0.065 0.050

1-1/4 1.375 0.065 0.055

1-1/2 1.625 0.072 0.060

2 2.125 0.083 0.070


more insulation may be required, particularly in southern climates, to prevent

surface condensation. Thus, at least 2 in. of insulation is recommended.

Because the pipe surface is cold, insulation must be applied before the

chilled water system is in operation to ensure a dry surface for application. Then,

the insulation must be finished with an outer jacket that includes a vapor barrier

to prevent moisture migration through the insulation and internal condensation.

The most common vapor barrier consists of a kraft paper and aluminum foil

laminate, reinforced with fiberglass threads (called foil scrim and kraft or FSK).

Preformed glass fiber insulation with a factory-applied all service jacket will

include the required vapor barrier.

3.2.2. Water Expansion and Air Removal

Water expands and contracts when heated and cooled in almost direct proportion

with the temperature change. Since water is incompressible, a lack of expansion

space in a closed piping system means that any volume increase will cause a

pressure increase that may result in mechanical damage to system components.

The most common way to accommodate water volume changes in a chilled water

system is the use an air cushion compression tank, as shown in Figure 3.2.

To determine the size required for a compression tank, Eq. (3.1) can be used.

V t ¼ ðEw 2 Ep 2 0:02ÞV s=½ðPa=PfÞ2 ðPa=PoÞ� ð3:1Þwhere Vt is the compression tank volume (gal); Ew 2 Ep, the unit expansion of the

system (typically 1% for chilled water systems); Vs, the volume of system (gal); Pa,

the atmospheric pressure (14.7 psia); Pf, the initial pressure in tank (usually

atmospheric pressure); Po, the final pressure or relief valve setting, psia (typically

100 psig or 114.7 psia).

The system volume, Vs, can be computed by adding the piping volume to

the volume of water in the chiller(s) and coils. The water volume in the piping can

be computed using the data from Table 3.4 and the lengths of each size of pipe in

the chilled water system. The manufacturer(s) can provide chiller and cooling

coil water volumes.

As water is heated or cooled, entrained air is released or absorbed in

proportion to its temperature and pressure. To remove air from chilled water

piping systems, an inline tangential air separator, as shown in Figure 3.3, is used.

In the separator, the water velocity is reduced to less than 0.5 ft/sec, which allows

the entrained air to separate from the water. The tangential flow pattern in the

separator creates a low pressure vortex in the center where the entrained air can

collect and rise through connecting piping to the compression tank above.

Air separators are sold with or without integral strainers. Since a strainer in

a piping system is designed to filter and hold solids (scale, dirt, etc.), the separator

strainer, coupled with the low water velocity, is the point in the chilled water

piping system where “sludge” will tend to collect. However, most owners are

Chapter 350

FIG

URE3.2.

Compressiontankinstallationandpiping.


unaware that their separator contains a strainer that must be routinely opened and

cleaned. This maintenance item is difficult and results in significant water loss.

Since there are other strainers in the system (usually located at the pumps)

that are designed for routine maintenance, a strainer in the separator is both

superfluous and an unnecessary maintenance burden. It is recommended that air

separators without integral strainers be used.

3.2.3. Water Treatment

Chilled water systems are “closed recirculating water” systems. For water

treatment, closed systems have many advantages.

1. There is no loss of water in the system (except when a leak occurs) and,

thus, no need for make-up water. Therefore, deposition or scaling is not

a problem.

2. Once filled and entrained air is removed, the closed system creates an

anaerobic environment that eliminates biological fouling as a problem.

3. Finally, closed systems reduce corrosion problems because the water is

not continuously saturated with oxygen (an oxidizer) as in open

systems. The low temperatures inherent with chilled water systems

further reduces the potential for corrosion.

TABLE 3.4. Average Water Volume in Piping

Pipesize(in.)

Outsidesurface area

(sf/lf)

Watervolume

(gallons/lf)

1/2 0.221 0.0163/4 0.275 0.0451 0.344 0.0781-1/4 0.435 0.1061-1/2 0.497 0.1742 0.622 0.2492-1/2 0.753 0.3843 0.916 0.5144 1.18 0.6616 1.73 1.5018 2.26 2.59910 2.81 4.09612 3.37 5.87514 3.93 7.16316 4.44 9.48918 4.71 12.0420 5.24 15.0124 6.28 22.11

Chapter 352

FIG

URE3.3.

Tangential

airseparatorinstallationandpiping.


Water treatment to eliminate any potential for corrosion is required only when the

chilled water system is initially filled with water, or when it is drained and re-

filled for maintenance, repair, or modification.

Corrosion in a closed system can occur due to oxygen pitting, galvanic

action, and/or crevice attack. To prevent these conditions, the “shot feed” method

of chemical treatment is used. With this method, a bypass chemical feeder, as

shown in Figure 3.4 is used to add treatment chemicals to the system in a one-time

“shot” just after the system is filled with water. For the mixed metallurgy (steel

and copper) systems typical for chilled water systems, a molybdate corrosion

inhibitor is best. Recommended treatment limits are 200–300 ppm molybdate.

3.3. PUMP SELECTION AND PIPING3.3.1. Pump Basics

Chilled water pumps are almost universally centrifugal pumps. A pump converts

the energy provided by a prime mover, most often an electric motor, to energy

within the liquid being pumped. This energy within the liquid is present as

velocity energy, pressure energy, static elevation energy, or a combination of two

or more of these. The rotating element of a centrifugal pump, which is turned by

the prime mover, is called the impeller. The liquid being pumped surrounds the

impeller, being contained by the pump casing, and, as the impeller rotates, its

motion imparts a rotating motion to the liquid, as shown in Figure 3.5. The entire

pump assembly is illustrated in Figure 3.6.

FIGURE 3.4. Bypass chemical shot feeder.

Chapter 354

FIGURE 3.5. Operation of a centrifugal pump impeller.

FIGURE 3.6. Centrifugal pump configuration.


Liquid enters the impeller at its center at low pressure and low velocity. As

the liquid travels outward along the vanes of the impeller, it acquires velocity and

an associated increased velocity pressure. But, when the liquid leaves the end of

the vanes, as shown in Figure 3.5, it slows rapidly and the majority of the velocity

pressure is converted to static pressure or pump head.

FIGURE 3.7. (a) Recommended end-suction (single suction) base-mounted

pump installation. (b) Recommended horizontal spilt case (double suction) base-

mounted pump installation. (c) Recommended vertical split case (double suction)

base-mounted pump installation.

Chapter 356

The centrifugal pumps used in water systems are typically as follows:

Line-mounted pumps can be installed directly in the piping since the

suction and discharge connections are arranged 1808 apart. The motor

and pump shafts, typically, are mounted vertically. The pump may be

supported by the piping and/or by additional hangers or a foot stand.

Based-mounted pumps have the motor and pump shafts mounted

horizontally, with both the pump and the motor bolted to a common

frame or base. These pumps are available in three configurations, as

illustrated in Figure 3.7: end-suction, horizontal split case, and vertical

split case.

Pumps are selected on the basis of the required flow and head, the pump

efficiency, and the application or location requirement. Line-mounted pumps

are normally limited to about 600 gpm maximum flow (about 75Hp), while

end suction pumps have a maximum flow of about 1000 gpm (about 60Hp).

There is essentially no flow limit for horizontal split case, vertical split case,

and vertical turbine pumps since these pumps can be fabricated as large as

required.

3.3.2. Pump Head and Horsepower

The required flow rate for the chilled water pump(s) is determined from Eq. (1.1).

The next step in defining the pump requirement is to compute the required pump

head, the pressure loss that the pump must overcome at the required flow rate.

FIGURE 3.7. Continued.


The pump head in a closed chilled water system is the sum of two pressure

components in the piping system:

1. Head loss due to friction in the piping system. Water piping is typically

sized utilizing Figure 3.8, based on a recommended design head loss

factor of between 3 and 6 ft of water/100 ft of piping and a maximum

flow velocity of 10 fps. Thus, a water flow of 1000 gpm would, from

Figure 3.8, require an 8 in. pipe size, resulting in a friction head loss of

1.5 ft of water/100 ft of piping. Additionally, a conservative approach

to account for losses in piping fittings and specialties, including elbows,

tees, valves, strainers, air separators, etc., is to use 50% of the straight

pipe pressure losses.

Thus, the friction head loss can be conservatively estimated from

Eq. (3.2).

H ¼ 1:5ðL=100ÞF ð3:2ÞwhereH is the head loss due to friction, ft of water; L, the piping length,

ft; and F, the friction head loss factor from Figure 3.8, ft of water/100 ft

of pipe.

2. The pressure loss through the chiller evaporator, cooling coils, control

valves, and piping fittings and specialties must also be included. For the

chiller evaporator and cooling coils, the pressure loss at design flow can

be provided by the manufacturer. Control valve losses are

approximately 5 psig or 13 ft (see Chap. 4).

The required brake horsepower for a water pump can be estimated using Eq.

(3.3).

BHP ¼ ðTHD £ FÞ=ð3960 £ EffÞ ð3:3Þwhere BHP is the pump brake horsepower; THD, the total pump head, ft. of water

column; F, the water flow rate, gpm; Eff, the (typical) pump efficiency: vertical

split case—0.81, horizontal split case—0.82, end suction—0.70, inline—0.70,

and vertical turbine—0.78.

3.3.3. Variable Flow Pumping

Variable frequency drives (VFDs) are applied to chilled pumps to create variable

flow pumping for primary–secondary and variable primary flow configurations.

Pump motor speed (rpm) is defined by Eq. (3.4):

RPM ¼ Fð120=PolesÞ ð4Þwhere RPM is the motor speed, revolutions/min; F, the electrical frequency,

0–60Hz; and Poles, the number of motor poles (2–8).

Chapter 358

FIG

URE3.8.

Frictionloss

forwater

inschedule

40commercial

steelpipe.


At full or synchronous speed at 60Hz, the motor rpm is defined as by

Table 3.5.

There are three major VFD designs commonly used: pulse width

modulation (PWM), current source inverter (CSI), and variable voltage inverter

(VVI). For HVAC applications, the vast majority of drives are the PWM type.

These drives are reliable, efficient, affordable, and readily available from a

number of manufacturers.

At 60Hz, AC electrical service produces a sine wave of voltage vs. time.

To vary motor speed, the PWM VFD simulates a sine wave to supply the motor

with various “time widths” of voltage that, when averaged together over the

length of the cycle, looks like a sine wave to the motor. This is illustrated in

Figure 3.9.

The typical PWM VFD consists of two major components: a rectifier

converts the 60Hz AC input to DC output to an inverter that converts the DC

back to AC output at a controlled frequency. For the motor to supply constant

torque, the output voltage is varied with the output frequency to maintain a

constant relationship between the two (7.6V/Hz for a 460V motor and 3.8V/Hz

for a 230V motor).

TABLE 3.5. Motor Synchronous Speed

Number of poles Synchronous speed (RPM)

2 3600

4 1800

6 1200

8 900

FIGURE 3.9. Pulse width modulation as a frequency control method.

Chapter 360

Efficiency of the PWM VFD is 92–96 þ %. The VFD contains a solid-

state reduced voltage motor starter and starts the motor at low speed to soft load

the system.

It is important that each VFD be provided with the following:

1. Main power disconnect ahead of drive—If the motor disconnect switch

is located between the drive and the motor, the drive can be destroyed if

the disconnect is closed with the motor already spinning. The VFD will

try to pull the motor down to a low, soft-start frequency, which can

result in high current and an overload on the drive. This means that the

VFD must be located close to the motor (to comply with National

Electric Code requirements).

2. Bypass switching—In the event of VFD failure, an integral bypass

switch, either manual or automatic, will allow the motor to be operated

at full speed.


4

Chiller Controls

4.1. START-UP CONTROL

There are two basic methods used for controlling the start-up of a chilled water

system:

1. The most common method is “manual” initiation. With this method, the

facility operating staff makes the decision to start the system on the

basis of the time of year, outdoor temperature, and/or the number of

“hot” complaints received from the building occupants. This method is

widely used in northern climates that have more distinctly separate

heating and cooling seasons.

2. Outdoor temperature can be the trigger for starting a chilled water

system, particularly if airside economizer systems are used in the

facility. With economizers, outdoor air will satisfy the cooling

requirement with the temperature is at or below the air-handling units’

discharge temperature setpoint, usually about 558F. This temperature,

then, is set as the “change-over” temperature and the chilled water

system is started with the outdoor temperature rises above this setpoint.

Starting a chilled water system means, first of all, starting the chilled water

distribution pumps. After that, the chiller will start under its internal controls if

the chilled water supply temperature is above its setpoint condition. First, a

62

water-cooled chiller will start its condenser water pump. Then, if flow of both

chilled water and condenser water is “proven” by flow switches (preferably,

differential pressure type), the chiller will start.

Aside from the flow switches, every chiller is equipped with basic

operating safety controls to protect the refrigeration machine from damage:

1. Low refrigerant temperature

2. Low chilled water temperature

3. High condensing pressure

4. Low oil pressure

In the event that any of these conditions exist, the chiller will shut down and

require manual investigation and reset before operation can resume.

For rotary compressor chillers, there are typically two additional safety

cutouts:

1. High motor winding temperature

2. Motor overload (high amps)

4.2. CAPACITY CONTROL

Modern water chillers are normally provided with digital electronic controls

designed and integrated by the chiller manufacturer. In addition to the operating

safety elements outlined in the Section 4.1, these controls are used to provide

capacity control based on maintaining a supply chilled water temperature

setpoint.

4.2.1. Refrigerant Flow Control

For the vast majority of water chillers, capacity control means controlling the

refrigerant flow rate through the evaporator. Depending on the type of

compressor used, several methods are applied:

1. Reciprocating Compressors—Primary refrigerant flow is controlled by

the expansion valve that responds to the temperature of the refrigerant

gas leaving the evaporator to maintain setpoint. The throttling action of

the expansion valve causes a pressure increase in the system and the

compressor must then reduce its refrigerant flow rate to maintain its

discharge pressure setpoint.

The most common methods of reducing refrigerant flow rate and

capacity in reciprocating compressors are by opening suction valves,

bypassing refrigerant gas within the compressor, or bypassing

refrigerant gas flow externally to the compressor.

With the first method, called unloading, external actuators can

hold the suction valves, on one or more compressor cylinders, open.

Chiller Controls 63

Thus, no compression can take place in these cylinders and the gas flow

rate through them is reduced to zero.

With the other approach, called hot gas bypass, a solenoid valve

on the high pressure discharge on one or more cylinders can open and

divert the refrigerant flow back to the suction side of the cylinder. This

effectively reduces the pressure differential or “lift” produced by the

cylinder and reduces the refrigerant gas flow rate from the evaporator.

Unloading occurs in distinct steps based on the number of

cylinders in each compressor and number of compressors in each

chiller. For a four-cylinder compressor, there are four stages (or steps)

of capacity control, 25–50–75–100%. If there are two compressors in

the machine, this yields a total of eight steps of capacity control.

Obviously, the greater the number of compressors and the greater the

number of cylinders in each compressor, the “smoother” the capacity

control line.

2. Rotary Screw Compressors—Since the rotary screw compressor is a

positive displacement compressor, suction throttling can reduce the

refrigerant gas flow into the compressor. A modulating control method

is desirable in order to produce essentially infinite capacity adjustment

between the minimum and maximum flow rates and capacities.

A slide valve is a hot gas bypass control valve with a sliding action

arranged parallel to the rotor bores and located at the high pressure

discharge of the compressor. The valve is then modulated to return a

variable portion of the discharge gas back to the compressor suction.

This valve, in addition to controlling capacity, also adjusts the location

of the compressor discharge port as the load changes. This “axial

discharge port” then provides good part load performance without

reducing full-load efficiency.

3. Centrifugal Compressors—Refrigerant gas flow into a centrifugal

compressor can be controlled by adjustable inlet guide vanes, or

preswirl or prerotation vanes, just as with a centrifugal fan. These

vanes are arranged radially at the inlet to the compressor impeller and

can be opened and closed by an external operator.

Since each vane rotates around an axial shaft, they affect the

direction of the flow entering the impeller. When the inlet vanes are

fully open, gas enters the impeller at 908 to the impeller. However, as

the inlet vanes begin to close, flow enters the impeller at an increasing

angle in the direction of the radial flow along the impeller blades. This

“preswirl” condition reduces the ability of the impeller to impart

kinetic energy to the refrigerant gas, thus reducing the flow rate.

Inlet vanes do not produce a pressure drop or “throttling” to

reduce refrigerant flow through the centrifugal compressor.

Chapter 464

A minimum volumetric rate flow through a centrifugal

compressor is required for stable operation. If the volumetric flow

rates falls below this minimum, the compressor will become unstable

and surge. When this happens, the refrigerant flows alternatively

backwards and forwards through the compressor, producing noise and

poor operation. Extended operation under surge conditions will cause

mechanical damage to the compressor. The surge envelope will vary

from compressor to compressor but usually occurs when the volumetric

flow rate is reduced by 40–60%.

To prevent surge from occurring, internal hot gas bypass may be

used to allow capacity to be reduced while maintaining sufficient gas

flow through the compressor. (This condition, along with increasing

windage losses and motor inefficiencies, accounts for the part load

performance characteristics shown in Fig. 1.7.)

In recent years, capacity control of centrifugal chillers by speed

control has been applied. Here, a large variable frequency drive is

applied to the chiller motor and the motor speed modulated to control

capacity. Generally, speed control improves efficiency over inlet vane

control down to about 55% of rated capacity, whole inlet vane control

is more efficient below 55% of rated capacity.

Compressor speed is directly related to capacity, but the pressure

(lift) produced by the compressor is a function of the square of the

speed. This may produce unsatisfactory operation and surge, requiring

the use of hot gas bypass, as the chiller unloads under speed control.

Adding speed control significantly increases the price of the

chiller and this option must be carefully evaluated to determine if this

concept is cost-effective for a given cooling load profile.

4.2.2. Sequencing Multiple Chillers

Where multiple chillers are utilized, these chillers must be sequenced on and off

to meet the imposed chilled water system load efficiently. To prevent rapid

stopping and starting, called short cycling, a time delay of about 30min is

required between the stopping and starting of an individual chiller. Additionally,

to equalize the run times for multiple chillers, a lead-lag or “rotation” sequence

should be implemented to change the order of stopping and starting the chillers

from one time to the next.

1. For constant flow chilled water systems, the imposed cooling load is

indicated by the return chilled water temperature, assuming the supply

chilled water temperature is maintained at setpoint.

For example, consider a system with a 448F supply chilled water

temperature, a 128F range, and three equally sized chillers arranged in

Chiller Controls 65

parallel. When the return water temperature is at 568F (or higher), all

three chillers are required. However, as the imposed cooling load

decreases, the return water temperature will also decrease and, when

the return temperature falls to 528F, one chiller can be stopped. As the

load continues to decrease and the return temperature falls to 488F, asecond chiller can be stopped.

As the cooling load increases, the chillers can be started at these

same temperature conditions.

2. With primary–secondary chilled water systems, the flow in the

secondary loop varies in proportion to the imposed cooling load since

the temperature range in this loop is maintained at a fixed setpoint.

Flow in the primary loop varies with the number of chillers and primary

chilled water pumps in use.

As discussed in Chap. 2, the flow rate in the primary loop must

always be equal to or greater than the flow in the secondary loop to

prevent blending of system return water flow with the supply water

flow, resulting in an elevated chilled water supply temperature. Thus,

control of the chillers stopping and starting can be based on the primary

return loop temperature downstream of the common piping or bridge.

Again, consider a system with a 448F supply chilled water

temperature, a 128F range, and three equally sized chillers arranged in a

primary–secondary configuration. At full cooling load, the flow rates

in the primary and secondary loops are exactly equal and the

temperature in the return section of the primary loop is equal to the

return temperature from the secondary loop, 568F.However, as the cooling load falls, the secondary flow rate

decreases. Since the primary loop flow rate now exceeds the secondary

loop flow rate, chilled water from the supply side of primary loop mixes

with return water from the secondary loop to produce a reduced

temperature in the return side of the primary loop. When this

temperature falls to 528F, one chiller can be stopped.

When a chiller is stopped, the flow rate in the primary loop is

reduced by one-third and now again equals the secondary flow rate.

Blending is eliminated and the temperature in the return side of the

primary loop returns to 568F.If the cooling load continues to fall, the temperature in the return

side of the primary loop will again began to fall as blending is

reinitiated. When it falls to 528F, the second chiller can be stopped.

As the cooling load increases, the flow in the secondary loop will

begin to exceed the flow in the primary loop. Under these conditions,

the temperature in the primary loop will rise to above 568F, whichindicates that a chiller must be started.

Chapter 466

The general algorithm for control of chiller sequencing in a

primary–secondary system made up of any number of equally sized

chiller modules (a chiller and a primary chilled water pump) can be

stated as follows:

T1 ¼ Tc þ Td, where Tc ¼ the system supply chilled water

temperature setpoint and Td ¼ the system design tempera-

ture range.

If (Tr þ 0.5), where Tr is the primary loop return water

temperature, is greater than T1 for at least 30min, start a

chiller.

T2 ¼ Td/N, where T2 is the range assigned to each of N equally-

sized chillers.

If (Tr 2 0.5) is less than (T1 2 T2) for at least 30min, stop a

chiller.

3. Sequencing chillers in a variable primary configuration is relatively

complex. Each chiller must be equipped with a high quality flow meter

that is used to measure flow through each chiller evaporator.

Again, consider a system with a 448F supply chilled water

temperature, a 128F range, and three equally sized chillers. At full load,

all chillers are operating at full flow, chilled water is supplied to the

coils at 448F and the return temperature is 568F.As the cooling load falls, the pump speed on each chiller reduces

to deliver less flow and maintain the return water temperature setpoint

of 568F. Once flow through all three chillers is reduced to 67% of

design flow, one chiller and pump can be stopped. Once the flow

through the remaining chillers is reduced to 50% of design flow, the

second chiller can be stopped.

As cooling load increases, the chillers can be started at these same

flow conditions.

4.2.3. Chilled Water Supply Temperature Reset

Chillers use less energy as the supply chilled water temperature is indexed

upward. Control of this indexing must, however, take into account both sensible

and latent cooling loads as independent variables:

1. Sensible Cooling Load—Chilled water supply temperature setpoint can

be indexed upward in the same proportion that the sensible load

decreases.

2. Latent Cooling Load—However, often the sensible cooling load will

reduce faster than the latent cooling load. If the chilled water

Chiller Controls 67

temperature setpoint is indexed upward solely on the basis of decreased

sensible cooling load, space humidity may increase to unacceptable

levels.

To properly set the chilled water supply temperature setpoint,

temperature and humidity conditions in each control zone must be

evaluated and the chilled water setpoint indexed upward only while

both conditions remain acceptable.

4.3. VARIABLE FLOW PUMPING CONTROL

Variable flow pumping is used in both primary–secondary and variable primary

flow configurations of chilled water systems. The starting point of variable flow

control is the cooling coil control valve. This must be a 2-way valve that throttles

flow through the coil as the cooling load decreases.

Water coils are arranged for counterflow configuration and the final

temperature of the leaving chilled water may exceed the outlet temperature of the

air leaving the coil. This is particularly important for chilled water systems that

have small temperature-difference requirements. Another advantage of the

counterflow configuration is that less surface area is required for a given rate of

heat transfer.

The overall heat transfer by the coil,Q, is determined by the mass flow rates

of the two media, the surface area, the temperature gradients, and the overall

coefficient of heat transfer, U, as defined by Eq. (4.1).

Q ¼ 2MhCp;hdTh ¼ McCpc dTc ¼ U dAðTh 2 TcÞ ð4:1Þwhere M ¼ mass flow rate, T ¼ temperature of the fluids, hot (h) and cold (c),

respectively, dA ¼ differential area of the coil under study.

Integrating Eq. (4.1) over the entire area of the coil yields

Q ¼ UAðTDi 2 TDoÞ=lnðTDi=TDoÞ ð4:2Þwhere TD ¼ Th 2 Tc at the inlet (i) and outlet (o) conditions, respectively. The

temperature term in Eq. (4.2) is called the logarithmic mean overall temperature

difference (LMTD).

Since U is defined by the design of the coil and the materials used in

construction, which is fairly consistent between manufacturers, the basic

performance of any coil is defined by the coil surface area and the temperatures

required, or LMTD. If large temperature differences are available, as for most

heating applications, the required coil area is small. For small temperature

differences, as for chilled water systems, the coil area is large.

Water coil surface area is defined by the number of rows of tubes and the

spacing, or density, of the fins (in terms of fins/inch of coil width). Thus, a heating

Chapter 468

coil may have 1 or 2 rows and 8–10 fins/in., but a cooling coil will often have

6–8 rows and up to 16 fins/in.

The capacity of any water coil is controlled by modulating the flow of

liquid through the coil. However, as flow is reduced, the capacity of the coil is not

reduced linearly. To compensate for this nonlinearity, equal percentage

characteristic control valves are utilized, resulting in a linear relationship

between coil capacity and valve stroke or percent open, as shown in Figure 4.1.

These control valves must be selected on the basis of the maximum design

flow through the coil and a “wide open” pressure drop equal to or greater than the

pressure drop through the coil. Additionally, the valves must be rated for tight

shutoff against the maximum system pressure that is developed, which, for

variable flow systems, is the pump “cut-off” pressure, the pressure the pump will

develop at zero flow.

As load is reduced at a cooling coil, the control valve throttles flow to the

coil to match flow capacity to the coil load, as follows:

1. Secondary chilled water pumps in a primary–secondary configuration

can reduce their flow rate by either increasing head (forcing a flow

reduction along the pump curve) or decreasing pump speed. The second

method is more efficient and produces significant energy savings

compared to the first. Thus, except for very small systems (less than

10 hp), VFD’s are used to modulate pump speed in response to flow

needs.

FIGURE 4.1. Coil capacity control with equal percentage control valve.

Chiller Controls 69

There are two methods to control pump speed:

A pressure sensor, located sufficiently downstream from the

pump to be sensitive to control valve action, can

respond to the pressure increase produced by

throttling control valves by reducing the pump speed

and, consequently, flow rate and pressure.

Pressure control represents an attempt to respond to a

symptom rather than the disease. While pressure does

change with flow changes, the change is not linear

(pressure changes as the square of flow), multiple

valves of different sizes may have different pressure/

flow relationships, and the resulting pressure change

in a larger system may have not real relationship to the

cooling load. The basis of design for a primary–

secondary system is maintain a fixed chilled water

supply temperature and a fixed temperature range in

the secondary loop. If, by chiller capacity control and

proper sequencing of chillers, the chilled water supply

temperature is maintained at setpoint, then the design

range can be used to establish a return temperature

setpoint to be used to control pump speed.

2. Variable flow control in a variable primary configuration is essentially

the same as for the secondary pumps in a primary–secondary

configuration.

As a pump speed is reduced, both the flow rate and pressure are reduced.

Ultimately, the available pressure at requested flow rates may not be sufficient to

ensure water flow to the more remote coils and, thus, these coils are “starved” and

they fail to cool. To prevent this problem, one or more differential pressure

sensors must be installed across the most remote flow circuit(s) and a minimum

pressure setpoint determined for each circuit. Then, as the pump speed is reduced,

if the pressure falls below the required setpoint in any circuit, further reduction of

pump speed in inhibited. At this point, the system becomes essentially constant

flow.

For variable primary flow systems, there is also a minimum flow

requirement through the chiller, equal to 3 fps or about 30% of design flow. A

bypass control valve must be modulated open when the flow rate through the last

operating chiller is reduced to its minimum level, as indicated by the chiller flow

meter.

Chapter 470

5

Thermal Storage

5.1. ECONOMICS OF THERMAL STORAGE

The goal for thermal storage in a chilled water system is to reduce the demand for

purchased cooling energy during on-peak periods by shifting cooling energy

consumption to off-peak periods. This reduces electrical charges, particularly

demand charges, on a year around basis.

Aside from shifting of cooling energy consumption from the on-peak to the

off-peak period, there may be some reduction in overall cooling energy

consumption with thermal energy. Every thermal energy system will have losses.

Water and ice storage systems have heat gains that increase the cooling energy

consumption. Water systems have thermal mixing problems that require more

cooling be stored than can actually be used. Ice systems, due to their reduced

temperatures, require greater compressor energy than an equivalent water storage

system. All of these losses may or may not be offset by improved chiller COP due

to reduced condensing temperatures during off-peak periods when lower outdoor

temperatures prevail.

Control of the thermal storage system is critical to ensure that enough

cooling energy is stored during the off-peak period to offset the shortfall in chiller

capacity during the on-peak period. Conversely, it is a waste of energy to store

more cooling than can be used during the following on-peak period.

71

The analysis of thermal storage economics requires that an accurate peak

24-hr cooling load profile be developed. Basically, this consists of developing,

for the cooling design day, the outdoor temperature and imposed cooling load at

each hour. Then, the day must be divided into the on-peak and off-peak periods

defined by the applicable electric rate schedule.

The total daily cooling requirement must still be met by the chiller, but,

utilizing thermal storage, the chiller runs for 24 hr, not 12–14 hr. The chiller,

then, can be sized on the basis of the “split” between the energy required for

charging during the off-peak period vs. the energy consumed in discharging

during the on-peak period, as shown in Figure 5.1.

Charging represents the off-peak period production of cooling energy by

the chiller that is stored, while discharging is the use of that stored energy during

the on-peak period. The amount of cooling required during the on-peak

discharge, divided by the amount of cooling energy stored during the off-peak

charging, is defined as the storage efficiency, typically 0.95 or greater for ice

systems and 0.90–0.95 for water systems.

Selecting a chiller and the optimum amount of storage requires an

evaluation of all the cost factors under different chiller capacity/storage ratios and

selecting the combination of chiller and storage that yields the best economic

performance. To illustrate this process, consider the following example: A large

building has a peak cooling load of 500 ton and a design day cooling load profile

as shown in Table 5.1. The building cooling system is operated from 6:00 AM to

FIGURE 5.1. Typical design day chilled water load profile.

Chapter 572

7:00 PM, resulting in a total daily cooling requirement of 4,880 ton-hr, 3,495 of

which occur during the on-peak period defined by the electrical rate as 12:00

noon to 8:00 PM.

A conventional system without thermal storage, Option 1, would produce

4,880 ton-hr of cooling by using a chiller without any storage. If the average

performance of the chiller is 0.6 kW/ton based on 858F water entering the

condenser, the energy consumption for cooling, including 0.2 kW/ton for pumps

and the cooling tower, would be 3,904 kWh, 2,876 of which are consumed during

the on-peak period. The on-peak demand for the 500 ton chiller and its auxiliaries

is 400 kW.

The next possibility is to minimize the use of on-peak energy by doing all

cooling off-peak, Option 2. This is sometimes called the “full storage” option. Thus,

TABLE 5.1. Example Design Day Cooling Load Profile

Hour O.A. temperature Cooling load (tons)

1 78 System off

2 76 System off

3 74 System off

4 76 System off

5 77 System off

6 78 275

7 80 160

8 81 190

9 82 200

10 84 160

11 86 300

12 88 350

13 90 400

14 92 450

15 93 500

16 94 450

17 93 400

18 92 375

19 90 300

20 86 270

21 84 System off

22 82 System off

23 81 System off

24 80 System off

Total 4880

Total on-peak 3495

Total off-peak 1385

Thermal Storage 73

the chillermust produce 4,880 ton-hr, plus storage inefficiencies, over a 16 hr period

(8:01 PM through 11:59 AM). Using 0.95 storage efficiency, the chiller size will be

321 ton and the storage capacity required is 3,784 ton-hr. Lower condenser water

temperatures and resulting condensing temperatures during the off-peak periodmay

reduce this to an average off-peak energy requirement of 0.5. The resulting daily off-

peak cooling energy use, including 0.2 kW/ton for pumps and cooling tower, is

3,595 kWh. The on-peak energy consumption is reduced to only about 0.1 kW/ton

for pumping. Thus, an on-peak demand of only 50 kW is incurred, with on-peak

electrical consumption of only 400 kWh.

Option 3 is to “level the load” by selecting a combination of chiller capacity

and storage such that the chiller would operate over the entire 24-hr design day at

essentially full load. For each hour, the combination of chiller and storage must

be sufficient to meet the imposed cooling load. If, during any hour, the chiller

produces more cooling than that required to meet the imposed load, the additional

cooling is stored for use later in the daily cycle.

Under this scheme, the chiller must produce the total cooling requirement,

plus storage inefficiencies, over a 24-hr period. In this example, the total cooling

requirement is 4,880 ton-hr. A chiller sized to meet this capacity, plus storage

inefficiencies, would be 214 ton, with a storage capacity of 2,018 ton-hr. As

shown in Table 5.2, the system performance results in on-peak demand and

consumption of 171 kW and 1,368 kWh, respectively, while off-peak demand

and consumption are 171 kW and 2,266 kWh, respectively.

The results of these three options are summarized in Table 5.3. The

consumption and demand charges for each option would have to be evaluated and

computed on an annual basis.

Another option that may be considered in locations with extremely high on-

peak demand charges is the “demand limiting” option. With this option, storage is

oversized so that the chiller can be operated at a reduced capacity during on-peak

hours.

The next step in the analysis for each of these cases is to determine the

capital costs for the chiller, cooling tower, pumps, piping, etc. Then, using simple

payback (or a more sophisticated analysis), the economic performance of each

option would be evaluated to determine the most cost-effective approach.

In general, the full storage option results in not only the greatest capital

requirement, but also the greatest savings in electrical costs. The load leveling

option reduces the capital requirements by minimizing both chiller and storage

capacity. The demand limiting option has demand savings greater than the load

leveling option, but lower than the full storage option.

There are economic risks associated with thermal storage. If the utility rate

schedule changes, there may be a negative impact on the cost performance of

thermal storage and it is imperative that a long-term contract be negotiated with

the utility. Another economic risk is system failure during the peak period, which

Chapter 574

TABLE5.2.

Exam

ple

DesignDay

CoolingEnergyConsumptionBased

onLoad

LevelingOption

Hour

O.A.temperature

Coolingload

(ton)

Chillercooling

(ton)

Storagecoolinga

(ton)

Inputenergy

(kW/ton)

TotalkW

178

0214

2214

0.6

128

276

0214

2214

0.6

128

374

0214

2214

0.6

128

476

0214

2214

0.6

128

577

0214

2214

0.6

128

678

275

214

þ61

0.6

128

780

160

214

254

0.6

128

881

190

214

224

0.6

128

982

200

214

214

0.7

150

10

84

260

214

þ46

0.7

150

11

86

300

214

þ86

0.8

171

12

88

350

214

1136

0.8

171

13

90

400

214

1186

0.8

171

14

92

450

214

1236

0.8

171

15

94

500

214

1286

0.8

171

16

93

450

214

1236

0.8

171

17

93

400

214

1186

0.8

171

18

92

375

214

1161

0.8

171

19

90

300

214

186

0.8

171

20

86

270

214

156

0.8

171

21

84

0214

2214

0.7

150

22

82

0214

2214

0.7

150

23

81

0214

2214

0.7

150

24

80

0214

2214

0.7

150

Total

4880

5136

22018/þ

1701

1368on-peak,2266off-peak

Note:Bold

fontrepresentson-peakperiod.

a(2

)indicates

charging,(þ

)indicates

discharging.

Thermal Storage 75

can eliminate the demand savings for an entire year. This, however, is only a risk

if other, standby chillers are brought on line to provide cooling while the storage

system is repaired. Again, the contract with the utility should allow a limited

number of short-term demand increases with no significant rate penalty.

Utilities are experimenting with real-time pricing rate schedules,

particularly in states that have implemented deregulation of their electric

utilities. Under these rates, the cost of energy is based on the actual cost of energy

production and/or power purchases by the utility on an hour-to-hour basis. Some

authorities predict that these rates will damage the economic viability of thermal

storage systems. Other experts point out that these rates will only impact control

methods and require variable chiller use vs. storage use. Again, any method that

increases the utility load factor reduces net costs to the utility and these savings

can be passed on to users who contribute to achieving them.

5.2. AVAILABLE TECHNOLOGIES

There are three general thermal storage system technologies: water, ice, and

eutectic salts. These technologies can vary significantly in terms of size,

efficiency, reliability, and cost, as described in the following sections and

summarized in Table 5.4.

5.2.1. Chilled Water Storage Systems

Chilled water storage systems use the sensible heat capacity of water, defined by

its specific heat of 1 Btu/lb 8F. These systems operate at temperature ranges

compatible with both reciprocating and rotary compressor water chillers and are

generally the most cost-effective method for thermal storage in larger systems

(2000 ton-hr or more).

The lowest temperature at which chilled water can be stored in the liquid

state is 328F. However, from a practical point of view, water temperatures lower

than about 368F are difficult to produce with rotary compressor chillers. As a

TABLE 5.3. Comparison of Cooling Thermal Storage Options

Result Option 1 Option 2 Option 3

Chiller capacity, ton 500 321 214

Storage capacity, ton-hr 0 3782 2018

On-peak demand, kW 400 50 171

On-peak consumption, kWh 2876 400 1368

Off-peak demand, kW 280 280 171

Off-peak consumption, kWh 1028 3595 2266

Total consumption, kWh 3904 3995 3634

Chapter 576

TABLE5.4.

ComparisonofAlternativeThermal

StorageTechnologies

Storage

medium

Volume

(cf/ton-hr)

Storage

temperature

(8F)

Discharge

temperature

(8F)

Advantages

Disadvantages

Chilled

water

12–15

36–44

40–46

Can

use

existingchillers,stored

water

canbeusedforfire

protection

Largestoragevolume,higher

heat

gains,difficultto

maintain

supply

temperature

setpoint

Ice

2.4–3.3

32

34–36

Low

storagevolume,high

dischargerates,easy

to

maintain

supply

temperature

setpoint

PoorchillerCOPdueto

low

temperature

requirem

ents,

expansionproblem,poor

thermal

conductivityofice

Eutectic

salts

647

48–50

Can

use

existingchillers

Relativelylargestoragevolume,

highchilledwater

supply

temperature

Thermal Storage 77

result, the cooling storage capacity required is 12–15 ft3/ton-hr (about

100 gal/ton-hr). But, chilled water storage costs per unit volume decrease with

tank size and heat gains per unit volume also decrease as the tank size increases.

The most difficult problem associated with water storage is maintaining the

required chilled water supply temperature, particularly near the end of the

discharge cycle, since any mixing of return water with supply water results in an

elevated supply water temperature. To address this condition, several approaches

are available:

1. Multiple Tank Storage—This method keeps return water from mixing

with supply water. If there are n tanks, one tank is empty, making the

storage volume ðn2 1Þ times the individual tank volume.

During charging, the tanks are filled one at a time with chilled

water. During discharge, one tank at a time supplies chilled water and

only after it is empty does return water enter the tank.

The obvious drawbacks to this scheme are the space requirements

for multiple tanks, the reduction in storage efficiency ½ðn2 1Þ=n� byhaving one empty tank, and the high heat gains due to increased tank

surface area.

2. Storage Tank with Siphon Baffles—A single large tank may be divided

into multiple sections by adding internal siphon baffles, as shown in

Figure 5.2. Water is supplied from one end of the tank and returned to

the other. Supply and return water mixing is reduced by the baffles and

the difference in densities between the supply and return water.

Experience with these tanks has not been good and their use is not

recommended unless the exact tank configuration has been tested

before installation.

3. Stratified Water Storage—Stratification is a method of keeping warmer

return water from mixing with colder supply water in a single tank.

Water that is less dense tends to rise to the top of the tank, while denser

water tends to fall to the bottom. Thus, supply water is removed from

the bottom of the tank, while water is returned to the top of the tank.

However, the overall density difference between supply and

return water at typical chilled water temperature ranges is only about

0.03%, so these tanks must be designed very carefully to prevent forced

circulation, with the resulting mixing, as water flows into and out of the

tanks.

Another condition must be considered: the density of water is

greatest at 39.28F. Therefore, cooling the supply water during the

charging cycle to temperatures below 408F is not recommended since

there will be tendency for the colder water to rise, effectively

destroying the stratification.

Chapter 578

5.2.2. Ice Storage

Ice storage has two great advantages over chilled water storage. First, ice storage

represents a phase change process with water absorbing or releasing 144Btu/lb to

melt or freeze at 328F. Thus, ice storage systems require only about 20% of the

storage volume of an equivalent water storage system. The second advantage is

that ice melts at a constant temperature, making the delivery of a constant chilled

water supply temperature very simple.

The major disadvantages to ice storage are the poor energy efficiency of

the freezing process and the complexity of the ice storage systems themselves.

These systems have relatively high cost and are most cost-effective only in

smaller storage capacities (less than 1,000 ton-hr). In addition, due to the

low temperatures, reciprocating compressor chillers are typically required,

though multiple rotary compressor chillers arranged in a “cascade” can be

applied.

Additional drawbacks to ice storage are two problems associated with ice

production:

1. Water expands as it freezes, producing a highly destructive force that

must be addressed by the ice storage equipment. Ice also exerts a

buoyancy force that must be addressed by some types of ice production

systems.

FIGURE 5.2. Chilled water storage tank with siphon baffles.

Thermal Storage 79

2. Ice has poor thermal conductivity. As ice forms on an evaporator

surface, lower evaporator temperatures are required to freeze a greater

thickness of ice.

There are a number of ice production and storage technologies currently

available.

1. Ice Shedders—A simple ice storage system consists of an evaporator

located above a tank. This “ice shedder” system, as shown

schematically in Figure 5.3, is designed to form ice on the evaporator

surface, then by mechanical means or by defrosting, separating the ice

from the evaporator surface so it will fall into a tank. Chilled water can

then circulated through the tank to serve the building cooling load.

Ice shedders aerate the water in the tank by pumping it

continuously over the evaporator in an open configuration. Therefore, a

heat exchanger is usually required to help isolate the chilled water

system from the increased corrosion potential due to dissolved oxygen.

2. External Melt Ice-on-Coil—A simple approach to ice storage is to

insert evaporator coils into a water tank and freeze the water around the

coils during the charging cycle. During discharge, the water

surrounding the evaporator coil and its ice charge is circulated through

FIGURE 5.3. Ice shedder thermal storage system schematic.

Chapter 580

the chilled water system to serve the building-cooling load. This system

is illustrated schematically in Figure 5.4.

However, because of the poor thermal conductivity of water, there

are limits to the thickness of ice that can be formed on the coil surface. As

this thickness increases, the evaporator temperature must be reduced to

continue the freezing process and this, in turn, reduces the chiller COP.

Another potential problem is ice bridging. To improve efficiency by

reducing ice thickness, evaporator coils are closely spaced in the tank.

However, if all the ice is not melted during the discharge cycle, ice may

grow from one evaporator to another, bridging the separation and cutting

off flow around these evaporators, reducing discharge efficiency.

Overall, these systems have a lower overall efficiency than other

types of ice storage systems.

3. Internal Melt Ice-on-Coil—With this system, the freezing point of

chilled water is lowered by adding ethylene glycol. During the

charging cycle, the chilled water (sometimes called “brine”) is

circulated through evaporator coils in water-filled tanks, and ice is

formed on the coils’ outer surfaces. During discharge, the ice is

melted by circulating warmer return chilled water through the coils,

cooling the water.

FIGURE 5.4. External melt coil freezing thermal storage system schematic.

Thermal Storage 81

This system avoids the problems of the external melt system

and eliminates the potential corrosion problems associated with ice

shedders. There are numerous commercially available packaged

storage systems based on the internal melt concept on the market today.

To minimize the heat gain by the large number of tanks typically

used with this technology, the tanks must be well insulated. Often, it is

cost-effective to bury the tanks, leaving only their tops exposed, to

reduce temperature gradients across the tank wall.

The basic operating modes of this technology is illustrated

schematically in Figure 5.5.

FIGURE 5.5. Internal melt coil freezing thermal storage system schematic.

Chapter 582

4. Ice Capsules—Again, the freezing point of chilled water is lowered by

adding ethylene glycol. With this technology, a tank is filled with

sealed containers of water submerged in the chilled water. During the

charging cycle, the water-filled capsules are frozen by circulating cold

chilled water through the tank. During discharge, as warmer chilled

water is circulated through the tank, the ice melts, cooling the water.

The capsules are made of thin-walled polyethylene plastic and are

designed to survive the expansion of freezing water. However, if water

is kept still, as inside these capsules, it may cool to well below 328Fbefore it actually freezes. Therefore, to ensure complete freezing, the

capsules must be cooled to about 258F.The basic operating modes of this technology are the same as for

the internal melt storage system.

5. Ice Slurry—With this technology, propylene glycol or calcium

magnesium acetate (CMA) is added to water and the resulting brine

is circulated over a special evaporator. As the brine flows down a tall

evaporator surface that is at a temperature equal to the freezing point of

the brine, ice begins to form. A mechanical scraper then removes the

ice from evaporator surface so that a mixture or “slurry” of ice and

liquid is formed.

A heat exchanger is used to isolate the glycol solution from the

building chilled water system.

5.2.3. Eutectic Salts Storage Systems

The main disadvantage to ice storage is that ice freezes at a temperature much

lower than is required for most HVAC applications, forcing compressors to

operate at much lower evaporator temperatures and increasing energy

consumption because of the lower COP’s that result. The main advantage to

ice storage is reduced storage volume because of phase change heat transfer.

From the earliest days of thermal storage systems there has been an

ongoing search for a storage material that would change phase at temperatures

closer to those required by conventional chilled water systems. These materials

are called eutectic from the Greek for “good freezing.”

The best results for this technology so far have been achieved with hydrated

salts. With these compounds, heat is absorbed as the water molecules are

separated from the salt molecules, a chemical reaction that while not strictly a

phase change is called that. This reaction occurs at approximately 478F, whileeutectic salts that changed phase at 40–428F would be more ideal.

The advantage of eutectic salt technology is the reduced storage volume.

These systems require only about half the volume of water storage, but still more

than twice the storage volume of ice systems.

Thermal Storage 83

The disadvantage of this technology is having to deal with the high phase

change temperature, which is typically too high for chilled water systems

operating design loads unless latent loads are very low. In addition, the discharge

temperature is not totally uniform, so a real chilled water temperature of about

508F can be expected. But, this temperature are well below the normal chilled

water return temperatures (particularly for systems with a larger design range), so

the system can be used in a series configuration with the chiller, precooling the

return chilled water before it enters the chiller.

5.3. APPLICATION OF THERMAL STORAGE

The use of thermal storage may be economically attractive only if the following

conditions apply:

1. The peak cooling load is much higher than the average cooling load

over the design day, i.e., peak loads are of short duration.

2. The electric utility rate includes high demand charges, ratchet charges,

or high differential between on-peak and off-peak rates.

Owners or designers contemplating thermal storage as part of a chilled water

system must do more evaluation than that for conventional systems. These steps

are recommended to determine the feasibility of a thermal storage system:

1. Consider the owner’s ability to operate a chilled water system with

thermal storage. These systems require ongoing monitoring and

adjustment to work properly and efficiently. System failure can have

significant negative financial impacts.

2. Determine the cooling load profile, hour-by-hour during the peak

cooling period, in order to determine required chiller and storage

capacities. Then, using a detailed annual cooling load profile for

existing facilities or hour-by-hour computer modeling for new

facilities, compute the annual operating cost for the system. Compute

electrical costs using the exact rate schedule(s) to be applied to this

system.

3. Select the thermal storage technology to be used. Ensure that all

operating modes are fully defined by the design and incorporated into

the system controls.

4. Determine the capital requirements and anticipated economic

performance of the thermal storage system. While simple payback

methods can be applied, more sophisticated methods, such as rate of

return or life cycle cost analysis, should be applied.

5. Negotiate a long-term contract with the electric utility. This contract

should establish the specific rates to used and provide for exemptions

Chapter 584

from demand peaks caused by occasional system failure. The contract

period must be longer than the payback period for the system.

6. While the thermal storage system is being installed, throughout the

commissioning period and through the first year of operation, the

operating staff must be fully involved, trained, and monitored to ensure

successful system performance.

Thermal Storage 85

6

Special Chiller Considerations

6.1. NOISE AND VIBRATION

Air-cooled chillers and the cooling towers associated with water-cooled chillers

are installed outdoors, presenting numerous problems relative to performance

and community noise that result from poor placement and/or incorrect screening.

See Secs. 11.4 and 14.3 for discussion of these topics.

Water-cooled chillers are located indoors, which means that noise and

vibration created by the chiller and its associated pumps may become a problem

throughout the building. Therefore, careful attention to location and installation

of water-cooled chillers is required.

Water chillers have broadband sound levels with characteristic frequency

ranges depending on the type of compressor (reciprocating, centrifugal, or

screw). And, because some centrifugal chillers produce more noise under part

load operation, chiller noise must be analyzed for the entire load range.

Sound is the hearing sensation caused by a physical disturbance in the air.

Noise is simply objectionable sound.

Sound waves in air arise from variations in pressure above and below the

ambient pressure and can be caused by repetitive pulsation in an air stream (due

to rotating fan wheels or blades) and/or the cascading water. Some can hear sound

in the frequency range of 20–20,000Hz, but in reality, few can hear below 30Hz

or above 8000Hz.

86

Noise is measured in terms of the sound pressure level, measured in

decibels (abbreviated as “dB”). Decibel values for sound represent a logarithmic

scale, each 10 dB increase or decrease indicating a factor of 10 increase or

decrease in the sound level, as shown in Figure 6.1.

The ear is not equally sensitive to all sound frequencies. The sound pressure

level from two sound sources may be identical, but if the sources have different

frequencies, the “loudness” of one sound will be perceived higher than the other if it

is concentrated within the frequency ranges within which the ear is more sensitive.

Therefore, acoustics engineers have developed frequency weighing for sound-level

meters, and themost commonweighing standard is the “A-Scale,”which givesmore

“weight” to the middle octave bands of the sound being measured. Table 6.1

tabulates the A-scale frequency response or weighting factors.

Additionally, sound-level meters are usually provided with a method to

time-average sound over a selected time period. Otherwise, it would be difficult

to measure rapidly varying sounds.

To prevent the transfer of sound and vibration into occupied spaces, the

following recommendations should be implemented:

1. If feasible, locate the chiller(s) in a mechanical equipment building

separate from the occupied building(s). If the chiller room is integrated

into the building, the walls and ceiling must be designed to prevent

noise transfer. Masonry construction and, perhaps, interior acoustical

lining are required.

FIGURE 6.1. Representative noise sound pressure levels.

Special Chiller Considerations 87

2. All piping and conduit penetrations must be sealed with flexible

materials to stop noise and prevent the transfer of vibration to the

building walls and floors.

3. Mount all equipment on vibration isolators. Pumps can be isolated with

inertia bases and spring isolators as shown in Figure 3.7. If the chiller is

installed in a ground floor or basement location, it can be isolated with

cork-and-rubber pads designed to support the weight of the chiller

without fully compressing. If located in the upper floors of a building,

the chiller should be installed with spring vibration isolators as shown

in Figure 6.2.

4. Make sure that all piping and electrical connections to equipment are

made with flexible connectors.

5. For critical applications, chiller noise can be reduced by a sound-

absorbing enclosure. Constructed of movable panels, these enclosures

consist of lightweight panels with 1–4 in. of glass fiber sound

insulation installed around and over the chiller. To allow air movement

and access for servicing, the panels can be arranged with minimum 3 ft

overlaps to eliminate straight line noise paths.

Chillers are tested for noise level in accordance with ARI Standard 570. This data

can be used to evaluate chiller selection and installation requirements to

minimize noise impact.

6.2. ELECTRICAL SERVICE

Electrical service for electric-drive vapor compression water chillers almost

always consists of three phase, four wire service with the voltage to load ratios

indicated in Table 6.2.

TABLE 6.1. A-Scale Sound Pressure

Weighting Factors

Frequency (Hz) Weighting factor (dBA)

31.5 239.4

63 226.2

125 216.1

250 28.6

500 23.2

1000 0.0

2000 þ1.2

4000 þ1.0

8000 21.1

Chapter 688

FIG

URE6.2.

Chillerinstallationforupper

level

mechanical

rooms.


There are four methods of motor starting commonly used for electric-drive

chillers. For small compressors (15 tons or less), full voltage or across-the-line

starting is typically used. However, this type of starting imposes a starting or

inrush current that is about 6 times greater than the normal full load operating

current of the motor.

High current inrush and high starting torque can cause problems in

electrical and mechanical systems, so, for larger motors, some type of reduced

voltage starting is normally used. For reciprocating compressors, the part-

winding starting method is often used. For rotary compressors, wye-delta and

solid-state starters are used. Each of these methods has advantages and

disadvantages in their application.

6.2.1. Part-Winding Starters

Part-winding starting is done by first applying power to part of a motor’s coil

windings for starting and then applying power to the remaining coil windings for

running. The motor must be a part-winding motor with two sets of identical

windings that can be used in parallel. The windings then produce reduced starting

current and reduced torque when energized in sequence.

The start winding draws about 65% of rated motor locked rotor current and

develops about 45% of the normal motor torque. After about 1 sec, the second

winding is connected in parallel with the start winding and the motor runs with

full current and full torque.

6.2.2. Wye-Delta Starters

Wye-delta starters can be used only on wye-delta motors that have sized leads

that allow for the motor winding to be connected in either the wye configuration

or the delta configuration.

Wye-delta starting begins by connecting the motor leads in a wye

configuration for starting. With this configuration, the motor receives

approximately 58% of line voltage and develops approximately 33% of normal

current and torque. After a time delay (that is usually adjustable), the wye

contactors in the starter open, momentarily de-energizing the motor, and the delta

TABLE 6.2. Typical Electrical Voltages for

Water Chiller Service

Compressor input

(kW)

Typical service voltage

(3-ph, 4-wire, 60Hz)

0–15 200–230

16–1000 460–600

1001 þ 4160

Chapter 690

contactors close, re-energizing the motor in a delta configuration with full current

and full torque.

This type of transition from wye to delta configuration is called “open”

since there is a momentary period when no voltage is applied to the motor. A

higher cost version of this starter provides “closed” transition via another set of

contactors and a resistor bank to maintain voltage to the motor windings during

the transition period.

Wye-delta starters are called “electromechanical” starters since their action

includes both electrical aspects and mechanical contactors within the starter that

must open and close during the starting sequence.

6.2.3. Solid-State Starters

To provide reduced voltage starting without the potential operating and

maintenance problems of electromechanical starters, solid starting can be used.

This type of starter uses a silicon-controlled rectifier (SCR-state) that controls

motor voltage, current, and torque as the motor accelerates during the starting

sequence.

Starting current can be adjusted from about 25 to 75% of the full load value.

The primary advantages to this type of starting is that it is almost infinitely

adjustable (providing smooth, stepless motor acceleration), small in size, very

rugged, and has no contacts to fail. One other advantage is that, due to its smaller

size, the starter can typically be unit-mounted, while much larger wye-delta

starters must be installed separately and require wiring runs between the starter

and the chiller motor. The primary disadvantage to the solid-state starter is cost,

which may be two or three times that of wye-delta starters.

6.2.4. Motors

Motors for electric-drive compressors are classified as hermetic or open, as

follows:

Hermetic Motors: These motors consist of a stator and rotor directly

connected to a refrigeration compressor, both of which are sealed in a

common enclosure or casing. In this configuration, refrigerant flowing

from the evaporator passes through and around the motor as it enters the

compressor to provide motor cooling.

Open Motors: These motors are standard NEMA Design B motors. which

are mounted on the chiller and connect to the compressor via a shaft and

coupling arrangement.

The open motor arrangement has the advantage of easier accessibility for service

and, in the event of motor failure, the refrigeration circuit is not contaminated.

However, careful alignment of the motor and compressor shafts is required


(both initially and as an ongoing maintenance item) to prevent vibration and

mechanical damage.

With hermetic motors, the heat generated due to motor inefficiency (5% or

less of the rated motor kW) is rejected to the refrigerant and then to the

condenser. However, motor heat from an open motor is rejected into the chiller

room, requiring additional ventilation to limit heat build-up. Typically, the

maximum allowable chiller room temperature is set at 1048F. With a 958Fambient air temperature (design day), the required additional ventilation air flow,

in cfm, to offset motor heat can be computed by Eq. (6.1).

cfm ¼ ðfull load kWÞð0:05Þð3413Þ=½1:1 £ ð1042 95Þ� or

cfm ¼ 17:2 £ ðfull load kWÞ ð6:1ÞThe oil pump on an electric-drive chiller is a small load that uses full-voltage

starting, but must be served separately from the main compressor motor. To avoid

this problem, and the potential for errors during installation, it is recommended

that the chiller be specified to have a single-point electrical service connection.

Thus, only one wiring run from the power source to the chiller is required.

Absorption chillers have several small pump motors, each of which will

almost always be 5 hp or less. Again, single-point electrical service connection

should be specified so that the full-voltage starters and wiring between the motors

and starters is part of the chiller package.

6.3. CHILLER HEAT RECOVERY

The heat collected by the chiller during cooling must be rejected through the

condenser to a “heat sink” as discussed in Chapter 1. However, if there is a

simultaneous need for heat, then this heat can be recovered and utilized rather

than simply being rejected to the outdoors.

During the 1960s, natural gas was not available in many parts of the

country for commercial building heating and many all-electric HVAC systems

were designed. However, due to the high cost of electricity for heating, efforts

were made to use rejected condenser heat for this purpose, essentially creating

large-scale internal source heat pump systems. However, there were serious

drawbacks to these systems:

1. To be applicable, the building had to have a high ratio of interior zones,

which required cooling any time they are occupied, relative to

perimeter zones, which required heating or cooling primarily as a

function of outdoor conditions. Equally important, these interior zones

had to have a high level of internal heat gain due to lights, people, and

appliances.

Chapter 692

2. The normal chiller leaving condensing water temperature of 958F is

very low for heating applications. The problem is compounded at

reduced cooling load conditions since this temperature decreases in

direct ratio to the cooling load. To help this condition, heat recovery

chillers are designed to operate with leaving condenser water

temperatures as high as 1108F. This, however, significantly increases

the chiller energy input and resulting operating cost.

3. Electrical costs must be high. The economics of heat recovery vs. the

use of an airside economizer cycle to provide “free” cooling during the

winter must be carefully evaluated to determine if this type of heat

recovery is cost-effective.

Even in the 1960s, this type of heat recovery had marginal performance and

economic benefits. In addition, as soon as natural gas became available, these

systems were abandoned. Since the 1960s, the energy cost picture has only gotten

worse while chiller heat recovery performance has not improved, essentially

eliminating the internal source heat pump concept from consideration.

However, if there is a need for low-level (85–958F) heat during the coolingseason, then recovery and use of condenser heat should be evaluated. Typical

applications include the following:

1. Domestic hot water preheating can be applied as shown in Figure 6.3.

Municipal supply water temperatures can range from 65 to 858F during

FIGURE 6.3. Water-cooled chiller heat recovery schematic.


the summer. For buildings with high domestic hot water requirements

(hospitals, research laboratories, dormitories, etc.), preheating the

incoming water via a water-to-water heat exchanger will reduce hot

water heating costs.

2. Summer reheat loads can be met with condenser heat and the use of a

water-to-water heat exchanger as shown in Figure 6.3. Since no space

heating is required during the summer, the maximum reheat

requirement is to bring the air to room temperature, typically 758F.This temperature is well within the range of the condenser water

temperature available.

3. Process heating loads can be met with condenser heat if their thermal

level is below about 808F.

Chapter 694

7

Chiller Operation andMaintenance

7.1. CHILLER COMMISSIONING

Every chilled water system can be operated on the basis of meeting two goals:

Goal 1: Satisfy imposed cooling loads:

Maintain proper supply water temperature.

Maintain proper supply water flow in variable flow systems.

Goal 2: Minimize the cost of cooling:

Maintain chilled water temperature as high as possible, but still low enough

to satisfy latent cooling loads.

Maintain condenser water supply temperature as low as feasible.

Operate each chiller, to the maximum possible extent, in the 40–85% load

range.

To ensure that this type of operation is attainable, the chilled water system must

be properly installed and placed into service and its operation verified by test, a

process called commissioning.

The first step in this process is to verify that the chiller installation is correct

and in accordance with both the design and the manufacturer’s recommendations.

The design engineer should be retained to review the installation process and

perform a final inspection to ensure compliance with the design requirements.

95

Before the chiller can be started for the first time, the chilled water pump(s)

must be placed in operation and the chilled water flow adjusted to within^5% of

design rates. The basic steps required to place the chilled water pump in service

include the following:

1. Check pump installation, including mountings, vibration isolators and

connectors, and piping specialties (valves, strainer, pressure gauges,

thermometers, etc.).

2. Check pump shaft and coupling alignment.

3. Lubricate pump shaft bearings as required by the manufacturer.

4. Lubricate motor shaft bearings as required by the manufacturer.

5. Turn shaft by hand to make sure the pump and motor turn freely.

6. “Bump” the motor on and check for proper rotation direction.

The air-cooled condenser or condenser water system must be placed into

operation (see Sec. 15.1). System controls must be tested and their proper

operation confirmed.

Chillers should be initially placed into service only by the manufacturer’s

service technicians, a process often called factory start-up.

The following checklist can be used for verifying both chiller installation

and functional performance upon start-up:

General Installation Checklist:

1. Test and balance completed (CHW and CDW flows within 25 to

þ10% of design)

2. Observe no visible water or oil leaks.

3. Observe no unusual noise or vibration.

4. Chilled water piping insulation in good condition.

5. Pressure gauges, thermometers installed and operable.

6. Confirm that O & M manuals are onsite.

7. Confirm that training provided for operating staff.

Additional Installation Checklist for Absorption Chillers:

1. Confirm that the steam control valve installed and operating for indirect

fired chiller.

2. Confirm that chiller has been leak tested according to manufacturer’s

instruction.

3. Confirm that unit was evacuated properly before charging with

refrigerant and solution.

4. Confirm that distilled water was used for refrigerant charge and proper

amount of inhibitor added.

5. Confirm that the proper amount of lithium bromide was installed.

6. Confirm that the proper amount of octyl alcohol was installed.

Chapter 796

7. Test electrical service to each pump for proper voltage/phase.

8. Test operating amps for each pump.

Functional Performance Checklist:

1. Confirm that the factory start-up and tests were completed and the

chiller is ready to be placed into service.

2. Determine entering CHW temperature and compare to design.

3. Determine leaving CHW temperature is within 18F of setpoint.

4. Compute CHW range and compare to design.

Electric-Drive Chillers:

1. Determine that the chiller FLA is within 5% of design.

2. Test volts, phase-to-phase.

3. Compute voltage imbalance and determine that the maximum of a-b-c-

average/average is less than 2%.

Absorption Chillers:

1. Determine that the high generator temperature is within design range.

2. Test high/low generator solution level.

3. Determine that supply steam pressure is within 5% of design.

4. Test automatic over-concentration, dilution cycles, and dilution on

shutdown.

Water-Cooled Chillers:

1. Confirm that the cooling tower has been commissioned (see Sec. 15.1)

2. Determine the entering CDW temperature and confirm that it is within

18F of setpoint.

3. Determine leaving CDW temperature and compare to design.

4. Determine CDW temperature range and compare to design.

Air-Cooled Chillers:

1. Determine ambient air temperature.

2. Determine chiller head pressure and compare to rated value.

Operations and Control:

1. Confirm that chiller appears to meet load.

2. Confirm that chiller operates without alarm conditions or safety

shutdowns.

3. Confirm that chiller start sequence operates properly.

4. Confirm that multi-chiller staging sequence operates properly.

Chiller Operation and Maintenance 97

7.2. CHILLER MAINTENANCE

When establishing a maintenance program for chillers, maintenance managers

have three options:

1. Implement the program fully in-house.

2. Outsource the entire program.

3. Use a combination of in-house and outsourced functions.

One of the most important benefits of a program that uses in-house personnel is

the institutional knowledge of those systems. Maintenance personnel have been

working with those chiller systems for years, and they most likely know what

many of the existing maintenance problems are.

Another benefit of using in-house personnel is long-term quality.

Maintenance managers and in-house personnel are better able to focus their

attention and efforts on both short- and long-term requirements, up to the

expected life of the equipment. Outsourced programs tend to have a much shorter

focus, i.e., the length of the maintenance contract.

However, the cost of establishing a complete in-house program can be very

high. Managers have to arrange for the training of maintenance personnel on the

specifics of maintaining the chiller(s) installed in the facility, provide specialized

training for infrequent testing and servicing, and purchase specialized test

equipment for many of the inspection and maintenance activities that must be

performed.

Combination in-house and out-sourced programs typically assign the

routine and preventive maintenance tasks to in-house personnel, while

contracting is utilized for specialized activities that require a higher level of

expertise than is available in-house. This approach maintains in-house personnel

being actively involved in the maintenance, while outside experts assist them by

performing the more complex and infrequent tasks. This arrangement helps

preserve the institutional knowledge and typically is the most cost-effective

approach.

No matter which maintenance scheme is utilized, routine maintenance

requirements for electric-drive vapor compression cycle water chillers fall into

four categories: continual monitoring, periodic checks, scheduled maintenance,

and long-term maintenance, as follows:

Continual monitoring/visual inspection. The majority of chiller operating

problems and maintenance needs are discovered by visual inspection and the

monitoring of equipment operating parameters. Figure 7.1 is a form that can be

used to guide chiller visual inspection and to collect operating data every two

hours during the day. This data will give a complete “snapshot” of the running

conditions of the chiller and variations in data between observations can be a

prime indicator of operating problems.

Chapter 798

Over the long term, this data can be used to predict maintenance

requirements. This data, along with the data collected during periodic checks and

routine maintenance procedures, can be plotted against time so that a trend or

change in conditions can be identified. A decision on maintenance can then be

made from the trend.

FIGURE 7.1. Chiller data logging form.


Periodic checks. Periodic checks are specific tests that are performed

annually (or more frequently) to evaluate some aspect of the chiller performance

and include the following:

Leak testing (via the pressure method).

Purge operation (low pressure systems).

System dryness.

Lubricant level.

Lubricant filter pressure drop.

Refrigerant quantity or level.

Water flow rates. (Adjust the pumps to provide design flow and measuring

the pressure drop across the evaporator and condenser. Compare these

values to the manufacturer’s data to determine if internal scaling is

occurring.)

Expansion valve operation (for reciprocating compressor systems).

Regularly scheduled maintenance:

Condenser and lubricant cooler cleaning should be done at the beginning of

the cooling season or every six months if the chiller is in use year around.

This process has become more difficult since the heat exchanger tubes in

newer chillers have internal fins. (To facilitate this cleaning, piping

connections at the chiller should be flanged or use grooved piping

connections and the insulation be designed for removal and replacement.

The piping must be configured so as to not interfere with the tube

cleaning space.)

Evaporator cleaning on open systems is required for applications using air

washers. Here, the chilled water system is “open”, just as the condenser

water system is open, and the evaporator tubes should be cleaned at the

same time the condenser tubes are cleaned.

Calibration of pressure, temperature, and flow controls should be done

quarterly.

Inspect starter wiring connections, contracts, and action. Power

connections on newly installed starters may relax and loosen after a

month or so of operation. Tighten and recheck annually thereafter.

Safety interlocks devices, such as flow switches and pump starter auxiliary

contracts, should be checked annually for proper operation.

Dielectric motor testing should be done annually. This test will

identify failures in motor winding insulation. Other motor tests

should include test for imbalance of electrical resistance among

windings, imbalance of total inductance with phase inductances,

power factor, capacitance imbalance, and running amperage vs.

nameplate amperage.

Chapter 7100

Tightness of hot gas valve (as applicable) should be checked annually. If

the valve is does provide tight shutoff, replace it.

Lubricant filter and drier should be changed annually.

Laboratory analysis of lubricant should be performed annually during the

first 10 years of chiller life and every six months thereafter. This oil

analysis will define the moisture content (not to exceed 50 ppm), oil

acidity (maximum 1 ppm) that may indicate oil oxidation and/or

refrigerant degradation due to high temperatures, and metals or metal

oxides that indicate chiller component wear and/or moisture in the oil.

Vibration levels for open drive chillers should be checked annually.

Hermetic chillers should be tested every 3–5 years, more frequently as

the chiller ages. Vibration analysis is useful for detecting compressor/

motor alignment problems for open drive chillers and to evaluate bearing

wear, gear misalignment, and other dynamic problems in any chiller.

There are several different methods of vibration testing. In general,

accelerometers are suitable for gear box and bearing tests, while a

velocity type sensor is necessary to measure the low frequency vibration

associated with compressor motor imbalance, drive misalignment, and

sleeve bearing wear.

Valve and bearing inspection in accordance with manufacturer’s

recommendation.

Relief valves (both refrigerant and water) should be checked annually.

Disconnect the vent piping at the valve outlet and visually inspect the

balve body and mechanism for corrosion, dirt, or leakage. If there are

problems, replace the valve, do not attempt to clean or repair it.

Extended (long term) maintenance checks:

Eddy current (electromagnetic) testing of heat exchanger tubes should

be performed every 5 years (or more frequently as the chiller ages).

This testing will typically detect tube pits, cracks, and tube wear

(thinning). Any tube that is replaced as result of this testing should be

examined and cross-sections cut so the cause of the defects can be

evaluated.

Compressor teardown and inspection of rotating components should be

performed every 5 years. Visually inspect the compressor impeller

(centrifugal compressor) or rotors (screw compressor), capacity control

devices and linkage, and bearings. Replace seals and gaskets during

reassembly. (Keep detailed records of these inspections and use this

information to define future repair or replacement needs.)

Other components must be serviced, inspected, and/or replaced at the

intervals recommended by the chiller manufacturer.


The maintenance requirements for engine-drive vapor compression cycle

chillers are much the same as for electric-drive chillers except that the significant

maintenance requirements for the engine must be added to the program. Depending

on the type of engine and fuel, the manufacturer’s maintenance program must be

strictly followed to endure reliable engine operation and to attain the expected

engine life.

Absorption water chillers have three significant maintenance elements:

Mechanical components. Absorption chillers have refrigerant and solution

pumps and a purge unit that must be maintained. The pumps are usually

hermitically sealed and are cooled and lubricated by the refrigerant. On

an annual basis, these pumps and motors must be inspected and tested to

insure continued operation. Every 3 years, the pump seals and bearings

will require inspection and, as indicated, replacement.

On a routine (daily) basis, the operation of the purge unit must be checked

for both proper operation and excess operation, indicating an air leak. Air

leakage into an absorption chiller can produce serious problems due to

corrosion, contamination of the absorbent solution, and reduction in

efficiency and capacity.

Heat transfer components. Annually, the condenser and absorber heat

exchanger tubes must be cleaned. The lithium bromide solution must be

analyzed annually for contamination, pH, corrosion-inhibitor level, and

performance additive (octyl alcohol) and the solution adjusted as necessary.

Leak testing using the pressure method, is required annually.

Eddy current testing of the absorber, condenser, generator, and

evaporator should be done every 3–5 years.

Absorption chillers have a number of service valves, which contain

rubber diaphragms. These diaphragms should be replaced every

3 years (which requires that the lithium bromide/water solution

from the chiller).

Controls. Proper operation of controls is critical in preventing performance

problems with absorption chillers. Test the controls for proper operation

at the beginning of the cooling season. Clean and tighten all connections,

including field sensor connections. Vacuum control cabinets to remove

dirt and dust. With microprocessor controls, have the factory service

technicians test calibration and operation and ensure that the latest

version of operating software is loaded.

For direct-fired absorption chillers, the burner must also be maintained and

serviced to ensure efficient, reliable operation. The following is a summary of

basic burner maintenance requirements:

Annually:

Inspect and repair stack/breeching.

Chapter 7102

Clean heating surfaces.

Check combustion air intakes.

Test all safety controls, operating levels, etc.

Test relief or safety valve for operation and setpoint.

Routine maintenance:

Run efficiency test and adjust air/fuel ration at least twice each year (more

often for larger burners)

7.3. CHILLER PERFORMANCETROUBLESHOOTING

While both vapor compression cycle chillers and absorption chillers are reliable

machines that require little “hands-on” attention by their operators, they are

mechanical devices that are susceptible to failure and resulting operating

problems. When operating problems do appear, they generally fall within the

following three categories:

7.3.1. Selection or Design Problems

Typically, there are two common chiller problems that can traced to improper

design and/or incorrect chiller selection:

1. Capacity is too low or too high for imposed load. The first problem,

insufficient capacity, is fairly rare because most design engineers are

conservative folks who would rather err on the high side. But, simple

mistakes or changed conditions can result in a chiller that simply has

too little cooling capacity.

Unfortunately, there is no simple way of increasing capacity.

However, the problem can be mitigated by keeping condensing

temperatures as low as possible. Adding larger condenser coils for air-

cooled applications or an additional cooling tower cell can be far less

expensive than adding additional chiller capacity if the shortfall is

relatively small.

High head pressure in a chiller, which reduces capacity, is

normally the product of high condenser temperature conditions. This

lost capacity, also, can be regained by improving condenser side

performance.

Chiller over-capacity is a far more common problem, resulting in

inefficient performance and the potential for control point offset that

may cause the chiller to shut down due to low refrigerant temperature

or low chilled water temperature. Generally, this problem can be


addressed by false-loading the chiller with hot gas bypass. This will

prevent the chiller from going off-line, but will not improve energy use.

2. Chilled water supply temperature is too high. As discussed in Chapter

1, for an HVAC system to simultaneously maintain the desired space

temperature and humidity level, the chilled water temperature must be

low enough to allow supply air to be cooled to the required

temperature. If the chilled water temperature is too high, high space

humidity conditions will occur.

The chiller supply water temperature setpoint can be lowered from

the rated value, but there will be a corresponding reduction in chiller

capacity and increase in chiller energy consumption. Thus, there is a

trade-off betweenwater temperature and capacity thatmust be evaluated

to determine how low the supply water temperature can be set without

reducing capacity to below the imposed load.

Again, both chilled water supply temperature and capacity can be

improved by lowering the condenser side temperatures as much as

possible.

7.3.2. Installation Problems

Assuming that water flow rates are properly adjusted and controlled, controls

work properly, and the chiller is installed in accordance with the design

requirements, there are few operating problems. However, there is one common

problem in larger, primary–secondary systems that results from installation

problems with chilled water coils and control valves.

In most primary–secondary systems, the design chilled water range ðT r 2T sÞ for both the primary and secondary loops is the same—that is, coils and

chillers are typically selected for the same temperature rise. But, in many

systems, a low range problem shows up, as follows:

Symptom Having to operate more chillers than the load dictates in order

to have enough chilled water flow to meet the needs of certain

air-handling units.

Result System range ðT r 2 T sÞ is lower than design. Thus, chiller

plant capacity is actually “lost” due to the inability to obtain

design range

This lost chiller capacity is the problem since the condition requires thatmore chillers

be operated at lower load conditions. The operating conditionmay often be less than

about 30% of the chiller full load and, thus, chiller efficiency is lost. Thus, it simply

costs more to produce cooling, both in chiller operation and in pumping energy.

If the cooling coil heat transfer characteristics and the control valve flow

control characteristics are not exactlymatched, excess flow for the required part load

Chapter 7104

capacity will result. If the coil is dirty, the heat transfer factor becomes poorer,

requiring higher flow rates for a given part load capacity. If the control valve leaks,

excess flow results. All of these conditions may be contributors to the low range

problem, and all are difficult to find and fix in the field. However, there are a number

of steps that can be taken to improve the problem:

1. Clean all cooling coils on a regular basis to minimize fouling and

airflow resistance. If the coil is old and dirty, consider replacing it with

a new cooling coil selected for higher range, typically 25% higher than

the overall system design range.

2. Add sensible cooling only coils upstream of existing coils and pipe the

chilled water flow in series with the existing coil. Thus, a higher water

range can be obtained, with only a minor airside pressure drop penalty

(since these are typically 2–3 row coils).

3. Control valves must be properly sized 2-way equal percentage type.

(Make sure there are no 3-way valves in the system.) The valvewide-open

pressure drop at full flow should be (a) 1.25–1.50 times the coil pressure

or (b) equal to the excess branch circuit pressure differential, whichever is

greater. Specify 50:1 minimum rangeability. Valve operators must be

selected for tight shutoff against the pump cut-off pressure.

4. All control valves must be normally closed and arranged to close if the

air-handler is off (includes fan coil units). Bypass water flow can be a

root cause of the low range problem.

5. Raise (reset upward) Ts at reduced cooling loads (but, watch humidity

conditions).

6. Calibrate and maintain airside temperature controls. If these read

“high,” excess chilled water flow will result.

7. Monitor range. If you cannot measure it, you cannot manage it.

7.3.3. Maintenance Problems

Proper mechanical maintenance of the HVAC water chiller is required to keep it

performing as needed. (See Sec. 7.2.)

7.4. REFRIGERANT MANAGEMENT PROGRAM

Effective July 1, 1992, under Section 608 of Clean Air Act of 1990, the EPA has

established regulations (published in the Code of Federal Regulations, 40 CFR

Part 82, Subpart F) that require the following:

1. Practices that maximize the recycling of ozone-depleting compounds

(CFC’s and HCFC’s) during the operation, servicing, and disposal of

refrigeration equipment.


2. Certification of refrigerant recovery and recycling equipment,

technicians involved with handling refrigerants, and reclaimers.

3. Limiting of sale of refrigerant to certified technicians.

4. Repair of substantial leaks in refrigeration equipment with a refrigerant

charge greater than 50 pounds (equivalent to about 15 ton capacity or

larger).

5. Safe disposal requirements to ensure removal of refrigerants from

goods that enter the waste stream.

Vapor compression cycle water chillers using R-22, R-123, or other CFC/HCFC

refrigerants fall within the jurisdiction of these regulations. Absorption water

chillers and vapor compression cycle water chillers using R-134A, R-407A, or R-

410C are exempt from these requirements simply because these refrigerants have

no ozone-depleting characteristics.

For CFC/HCFC chillers, the certification of equipment and technicians, the

limiting of leaks, and safe disposal of equipment are required parts of a

refrigerant management program that must be implemented by the chiller owner.

Equipment Certification. Equipment used to evacuate refrigerant from a

chiller must be certified by an EPA-approved equipment testing organization

(either ARI or UL). Refrigerant removed from a chiller can be returned to that

chiller or to another chiller with the same ownership. However, if the refrigerant

changes hands, the refrigerant must be “relaimed” (i.e., cleaned to ARI Standard

700 purity level and chemically analyzed to verify that it meets this standard).

Technician Certification. EPA requires certification of persons who

perform maintenance, service, repair, or disposal that could be reasonably

expected to release refrigerants into the atmosphere. The definition of

“technician” includes anyone would attach/detach hoses and gauges to measure

pressure or would add or remove refrigerant from the system.

Four types of certification have been established by EPA, but for chiller

applications “Type II (high pressure),” “Type III (low pressure),” or “Universal

(any type of equipment)” certifications would be required. Technicians are

required to pass an EPA-approved test given by an EPA-approved certifying

organization.

Refrigerant Venting and Leaks. Small amounts of refrigerant released

during the purge of low-pressure systems are allowed by the regulations. Also,

small leaks when hoses or connections are disconnected, if meeting low-loss

standards, are also allowed. However, any chiller containing 50 pounds or more

of refrigerant must not leak more than 15% of the total charge in a year’s time. If

a leak is discovered (evidenced by the need to add refrigerant to maintain

capacity and/or prevent evaporator low temperature problems), owners have

30 days to repair the leak.

Chapter 7106

Disposal. A chiller that is removed from service must have all refrigerant

removed prior to disposal. This refrigerant can be retained for use in other chillers

under the same ownership or sold to a relaimer for reclamation and resale. (Note

that if refrigerants are recycled or reclaimed, they are not considered “hazardous”

by the EPA.)

The regulations also establish requirements for record keeping.

Technicians are required to provide owners with an invoice that indicates the

amount of refrigerant added to any system that contains 50 pounds or more. The

owner, then, is responsible for maintaining these invoices and using them to track

leakage rates from equipment. Additionally, owners must maintain servicing

records documenting the date and type of service, as well as the quantity of

refrigerant added.

Every organization that owns one or more chillers should review the

following checklist to establish and maintain a refrigerant management program:

1. Obtain copies of current EPA regulations and become familiar with

their requirements. Establish a program to collect, distribute, and

communicate updates and amendments that are issued regularly by the

EPA. (The EPA maintains a website at www.epa.gov that provides

significant information and links to specific documents that apply to a

refrigerant management program.)

2. Develop a “mission statement” specifically documenting your “intent

to comply” with EPA regulations.

3. Develop a written job description for a “Facility Refrigerant Manager”

who is responsible for refrigerant management and compliance with

EPA regulations.

4. Designate and train your “Facility Refrigerant Manager.” (This the

individual that the EPA will speak with when conducting a refrigeratn

compliance inspection.)

5. Develop written policy and procedures for refrigerant usage record

keeping, including specific method for collecting and maintaining

records and making them available to the EPA upon request.

6. Maintain copies of EPA certifications for all in-house and contracted

technicians working at the facility.

7. Develop written policies and procedures for each of the following

aspects of refrigerant handling and use:

Unintentional venting or leaking, including required reporting.

Refrigerant purchase requirements with a designated technician or

contractor responsible for all purchases.

Labeling for refrigerant cylinders.

Refrigerant inventory and storage, including in-out control and

auditing.


Leak testing process and service procedures for positive-pressure

chillers.

Disposal of refrigerant equipment, parts, and lubricants.

Refrigerant safety.

8. Develop a written “Emergency Response Plan” to address major

venting incidents, maximum exposure levels, and evacuation

procedures.

9. Include EPA regulation compliance language in the contracts with all

service organizations to ensure that they assume all liability in the event

of noncompliance.

10. Make sure that your refrigerant policies and procedures are “effectively

communicated to all affected personnel” through posting in the

workplace and by documented compliance training programs.

11. Annually, conduct a survey to ensure that your facility is in compliance

with EPA requirements.

To help in this process, the EPA publishes their Action Guide defining the

majority of compliance requirements. Also, there are consulting firms that can

provide detailed guidance in implementing a compliance program.

Chapter 7108

8

Buying a Chiller

8.1. DEFINING CHILLER PERFORMANCEREQUIREMENTS

The first step in purchasing a new or replacement chiller is to define the types of

chillers that are acceptable and the performance capacity for it.

As discussed in Chap. 2, the type of chiller can be defined as follows:

Vapor compression cycle, reciprocating compressor, electric-drive

Vapor compression cycle, rotary compressor, electric-drive

Vapor compression cycle, reciprocating compressor, engine-drive

Vapor compression cycle, rotary compressor, engine-drive

Absorption cycle, 2-stage, indirect-fired

Absorption cycle, 2-stage, direct-fired

To define capacity and performance requirements for any chiller, the following

parameters must be specified:

Maximum cooling capacity (tons)

Chilled water supply temperature (8F)Chilled water return temperature (8F)Chilled water flow rate (gpm)

Maximum evaporator pressure drop (ft wg)

109

Evaporator fouling factor (usually 0.00010)

Condenser water return temperature (8F)Condenser water supply temperature (8F)Condenser water flow rate (8F)Maximum condenser pressure drop (ft wg)

Condenser fouling factor (usually 0.00025)

Electrical service: V/ph/Hz

Maximum sound power level (dBA)

For electric-drive chillers, the motor requirements and allowable input energy

requirements must be specified:

Allowable motor type (hermetic and/or open)

Motor FLA

Maximum kW/ton at full load.

Maximum IPLV (kW/ton)

For engine-drive chillers, the engine parameters must be specified:

Engine type and fuel(s)

For indirect-fired absorption chillers, the steam or hot water parameters must be

specified:

Steam operating pressure (psig)

Maximum steam flow rate (lb/hr)

or

High temperature hot water temperature (8F)High temperature hot water flow rate (gpm)

For direct-fired absorption chillers, the burner parameters must be specified:

Burner type

Burner horsepower (hp)

Fuel(s)

Maximum firing rate (Btu/hr)

Minimum firing efficiency (%)

To compare the seasonal energy performance of alternative chillers, it is

necessary to (1) specify the load weighing factors and condensing conditions

to be applied to compute an NPLV and (2) require the chiller vendors to

provide the input energy or fuel requirement at 25, 50, 75, and 100% cooling

load for each alternative chiller. The weighing factors can be determined from

the cooling load profile (as discussed in Chap. 2), while condensing

temperatures can be estimated from weather data and the method of condenser

water temperature control.

Chapter 8110

8.2. ECONOMIC EVALUATION OF CHILLERSYSTEMS

The designer and/or owner should evaluate the total owning and operating cost of

each alternative chiller that will satisfy the performance requirements defined in

Sec. 8.1.

As shown in Sec. 2.3, chillers rarely operate for long atmaximumcapacity and

it is rarely sufficient to simply compare chiller peak load energy consumption in

order to make a purchase decision. The IPLV part load value, required by ARI

550/590-98 for electric-drive chillers, can be used to compare alternative chiller

seasonal energy consumptionusing a common ratingmethod.All other factors being

equal, the chiller with the lower IPLV will have lower seasonal operating costs.

However, as discussed in Sec. 2.4.3, the IPLV rating is only directly

applicable to the single chiller installation where the condenser water supply

temperature is allowed to fall as the outdoor wet bulb temperature falls.

Anytime multiple chillers are used, particularly if careful control allows the

chillers to be used in or near their optimum efficiency range of 40–80% of

rated capacity, the IPLV has little validity. Likewise, if control of the

condenser water temperature control method does not allow the supply

temperature to go as low as 658F, then the IPLV is not valid.

A variety of chiller components (tube quantity, shell length, compressor

size/quantity) can be selected by a chiller manufacturer in order to best satisfy a

given load profile. Because chillers can be customized for optimum performance

life cycle costing is the best way to compare alternative chiller selections. It

provides for a consistent method of analyzing the economic aspects of each

chiller and allows realistic comparison so that the most cost–effective, e.g.,

lowest life cycle cost chiller can be selected. The process is basically simple and

requires only that each potential chiller be evaluated using the same criteria. The

result is an “apples-to-apples” comparison, not “fruit salad.”

The computations required to determine life cycle cost utilizing the total

owning and operating cost methodology are simple. However, the

methodology’s accuracy depends wholly on the accuracy of the data utilized.

Anyone can calculate “garbage” life cycle costs simply because they use

data and/or assumptions that are “garbage.” Two different individuals, faced

with the same evaluation may compute wildly different life cycle costs because

they use significantly different data and/or assumptions in their computations.

The following subsections define the basic elements that make up the life

cycle cost.

8.2.1. First Costs

The initial capital costs associated with each potential chiller are all of the costs

that would be incurred in the design and construction of that chiller. Equipment

Buying a Chiller 111

costs can be obtained directly from the prospective equipment vendors.

Installation cost estimates can be obtained from local contractors or, lacking that,

from cost data published by R. S. Means Co., Inc. (Construction Plaza, 63 Smiths

Lane, Kingston, MA 02364-0800, 781/585-7880 or 800/334-3509).

The construction cost estimate must include the following, in addition to

the cost of the chiller itself:

Vapor compression cycle chiller ancillary costs:

1. Refrigeration room safety requirements (ASHRAE Std. 15)

2. Additional ventilation required for open motor.

3. Noise and vibration control.

4. Fuel piping, exhaust venting, and noise abatement for engine-drive

chiller.

Indirect-fired absorption cycle chiller ancillary costs:

1. Additional mechanical room area.

2. Additional floor slab or foundation structural requirements.

3. Additional condenser water system elements, including larger piping,

pumps, and cooling tower.

4. Back-up generator.

Direct-fired absorption cycle ancillary costs:

1. Same as for indirect-fired absorption chiller.

2. Additional mechanical room height.

3. Combustion air inlets.

4. Combustion venting (breeching, stack, etc.)

General and overhead costs:

1. Concrete housekeeping pad

2. Rigging

3. Demolition

4. Electrical power

5. Controls

6. Contractor Overhead (Insurance, bonds, taxes, and general office

operations, special conditions), typically 15–20%

7. Contractor Profit, typically 5–20%

Other costs that may be included in the capital requirement are design fees, which

may increase or decrease as a function of the selected alternative; special

consultants’ fees; special testing, etc. Also, chilled water and/or condenser water

piping may change configuration and cost between alternative chillers.

Unless the cost of the chiller is beingmet from operating revenues, at least a

portion of the capital expense will be met with borrowed funds. The use of this

Chapter 8112

money has a cost in the form of the applied interest rate and information about the

amount of borrowed funds, the applied interest rate, and the period of the loan

must be determined for the analysis. The total capital cost, including principal

plus interest, then can be computed using Eq. (8.1).

C ¼ Aþ {B £ n £ ½i=½12 ð12 iÞ2n�} ð8:1Þwhere C is the total capital cost, $, A is the first cost out-of-pocket, $; B is the

borrowed funds, $; i is the interest rate (decimal value); and n is the number of

years of loan.

8.2.2. Annual Recurring Costs

Once the chiller is placed into operation, two annual recurring costs must be met

each year of its economic life: energy costs and maintenance costs.

The economic life for an alternative is the time frame within which it

provides a positive benefit to the owner. Thus, when it costs more to operate

and maintain a piece of equipment than it would to replace it, the economic

life has ended. Typically, the economic life (sometimes called “service life”) is

the period over which the equipment is expected to last physically. The

economic life for reciprocating compressor chillers is normally selected as

15 years, while rotary compressors and absorption chillers have an expected

economic life of 23 years.

The computation of energy cost requires that two quantities be known: (1)

the amount of electrical energy consumed by the chiller and (2) the unit cost or

rate schedule for that energy. The second quantity is relatively easy to determine

by contacting the utilities serving the site or, for some campus facilities,

obtaining the cost for steam or power. That may be furnished from a central

source.

The energy cost, then, is computed by multiplying the electrical energy

consumption by the unit cost for electricity. Annual energy consumption is

computed from the cooling load profile (discussed in Chap. 2) and the part load

performance data provided by the vendors in accordance with Table 8.1 for each

alternative chiller.

Annual recurring maintenance cost is a very difficult element to estimate.

Lacking other information, the annual routine maintenance cost associated with

chillers can be estimated as 1–3% of the initial equipment cost.

8.2.3. Nonrecurring Repair and Replacement Costs

Nonrecurring costs represent repair and/or replacement costs that occur at

intervals longer than 1 year. For example, a compressor may require significant

repair (or even replacement) after 10–15 years of service. These costs must be

determined and the year of their occurrence estimated.

Buying a Chiller 113

8.2.4. Total Owning and Operating Cost Comparison

The total owning and operating cost for a chiller alternative and its associated

condenser water pump, over the system economic life, can be computed using

Eq. (8.2).

Life Cycle Cost ¼ C þ ðSumof Repair=Replacement CostsÞþ ½ðEconomic LifeÞ £ ðAnnual Energy Costþ AnnualMaintenanceCostÞ� ð8:2Þ

C in this equation is the total capital cost computed from Eq. (8.1).

8.3. PROCUREMENT SPECIFICATIONS

Recommended specifications for both reciprocating compressor and rotary

compressor water chillers are provided in Appendix B. Each specification

includes numerous options and, therefore, must be edited carefully before use.

TABLE 8.1. Chiller Utilization Profile

Load %

Load

(tons) Hours per year

Condenser water temperature entering chiller

(8F)

100

75

50

25

Chapter 8114

9

Cooling Tower Fundamentals

9.1. COOLING TOWERS IN HVAC SYSTEMS

As discussed in Chap. 1, in water-cooled HVAC systems, the cooling tower has

the role of rejecting the heat collected during the space cooling process to the

ambient atmosphere. In addition, the inefficiencies of the water chiller represent

heat that is added to the condenser water circuit and must be rejected by the

tower.

Once a design temperature range for the condenser water is established, the

condenser water flow rate can be determined from Eq. (9.1):

Fcdw ¼ ½Total Heat Rejection�=ð500 £ RangeÞ ð9:1Þwhere Fcdw ¼ condenser water flow rate, gpm; Total Heat Rejection ¼ cooling

system load þ heat of compression, Btu/hr; and Range ¼ condenser water

temperature rise, 8F.The total heat rejection value for Eq. (9.1) can be determined from the

compressor power input, as shown in Table 9.1.

As the condenser water flows through the tower, this heat is rejected to the

ambient air and the condenser water is cooled, primarily through evaporation of a

small percentage of the total water flow. Evaporation is a process by which heat is

absorbed by air and the remaining condenser water is cooled to the desired

leaving temperature.

115

All of the data presented inTable 9.1 is based on the use of electric-drive vapor

compression cycle systems.With absorptioncycle chillers, the amount of heat that is

added to the cooling load, and which must be rejected through the cooling tower, is

approximately 20–50% greater than for an electric-drive chiller. Therefore, the

condenserwater flowrate, for the same range,will be approximately 20–50%higher

(or, for the same flow, the range must be approximately 20–50% greater).

9.2. CONDENSER WATER SYSTEM ELEMENTS

The four elements of any condenser water system (discussed in Chap. 11) include

the following:

Condenser: The condenser is usually an integral part of the water chiller

(though separate refrigerant condensers are sometimes used). Typically,

the condenser is a shell-and-tube heat exchanger with the refrigerant in

the tubes, surrounded by condenser water in the shell. Heat transfer takes

place across the tube wall.

Condenser Water Pump: The pump circulates condenser water from the

cooling tower basin, through the piping system and the condenser, finally

returning the water to the cooling tower wet deck.

Condenser Water Piping: The piping for condenser water transport is

between the cooling tower and the condenser.

Cooling Tower: A device designed to reject heat from the condenser water

to the ambient air.

Every cooling tower, no matter how configured or constructed, must have the

elements shown in Figure 9.1 (and discussed in Chap. 10).

Fill: Heat transfer media in the cooling tower. Fill is designed to maximize

the “contact” between the return condenser water and the ambient air.

TABLE 9.1. Refrigeration System Total Heat of Rejection

Compressor

input power

(kW/ton)

Approximate heat of compression

(%), including 2% for condenser water pump(s)

Total Heat Rejection

(Btu/ton)

0.8 25 15,000

0.7 22 14,640

0.6 19 14,280

0.5 16 13,920

0.4 14 13,680

Chapter 9116

The better the contact, obviously better evaporation and heat transfer

occurs.

Hot Water Distribution (“Wet Deck”): Pans or basins with metering outlets

or spray nozzles designed to provide an even distribution of the return

condenser water entering the fill.

Cold Water Basin: The basin, either as an integral part of the tower or as a

separate sump, collects the water passing through the tower for supply to

the system by the condenser water pump. The basin must also be sized to

contain enough water to supply the condenser water system until the

pump returns water to the tower.

Fan(s): All cooling towers used for HVAC applications aremechanical draft

towers that use one or more fans to provide airflow through the tower.

Inlet Louvers and Drift Eliminators: Inlet louvers act to force the

air entering the tower into as straight and even flow pattern as possible,

while the drift eliminators are designed to trap and remove any entrained

water droplets that may be in the tower’s leaving air.

FIGURE 9.1. Cooling tower elements.

Cooling Tower Fundamentals 117

9.3. NOMENCLATURE

As shown already, cooling towers and condenser water systems have their own

vocabulary that must be mastered, as follows:

Approach: Difference between the condenser water supply temperature and

the entering wet bulb temperature, 8F.Blowdown or Bleedoff: Condenser water purposely discharged from the

system to control concentration of solids or other impurities in the water.

British Thermal Unit (Btu): Amount of heat required to raise or lower the

temperature of one pound of water by 18F.Capacity: Total amount of heat that the cooling tower can reject at a given

flow rate, approach, and ambient wet bulb temperature.

Casing: Exterior tower walls, excluding the louvers.

Cell: Smallest tower subdivision that can operate independently. Usually, a

tower cell is designed with independent water and airflow and cells are

controlled individually as a function of imposed overall cooling tower

load.

Counter Flow: Water flow path is configured at 1808 to the airflow path,

i.e., the two flows are in opposite directions.

Crossflow: Water flow path is configured at 908 to the airflow path, i.e., the

two flows are perpendicular to each other.

Double-Flow: A crossflow tower where two opposed fill banks and air

intakes are served by common fan and air plenum. (Called a “twin-pack”

in Great Britain.)

Drift: Unevaporated water droplets that are lost from the cooling tower.

Eliminators: An assembly of baffles or other devices to remove entrained

water droplets from the air leaving the cooling tower.

Evaporation (Loss): Condenser water that undergoes a phase change from

liquid to vapor and exits the tower as part of an air–vapor mixture.

Fill: Heat transfer media or surface within the tower, designed to maximize

the air–water contact surface area.

Float Valve: A make-up water control valve activated by a mechanical float

mechanism.

Forced Draft: Airflow is “pushed” through the cooling tower by one or more

fans at the air inlet(s), resulting in the tower being under positive pressure.

Head: Pressure required to be developed by the pump to overcome pressure

losses due to friction through the condenser water piping and the

condenser itself, the static lift required at the cooling tower, and any

residual pressure required by the cooling tower nozzles feet of water

column.

Chapter 9118

Induced Draft: Airflow is “pulled” through the cooling tower by one or

more fans at the air outlet(s), resulting in the tower being under negative

pressure.

Inlet Louvers: Blades or other assemblies designed to (1) prevent water

from splashing out of the tower and (2) promote uniform airflow through

the fill.

Lift: Static head represented by the height (in feet) between the water level

in the cooling tower basin and the wet deck water entry point, feet of

water column.

Make-Up (Water): Water added to offset water lost to evaporation and drift.

Mechanical Draft: Airflow through the tower is produced by one or more

fans.

Natural Draft: Airflow through the tower is produced by air density

differences produced by air temperature rise, i.e., chimney effect.

Nozzle: Device to control water flow through a cooling tower. Normally,

nozzles are designed to produce a spray pattern by either pressure or

gravity flow.

pH: Measurement of the acidity or alkalinity of condenser water.

Plume: Water vapor and heated air mixture discharged from a cooling

tower. (When condensation occurs, the water vapor becomes visible and

this is sometimes considered objectionable.)

Psychrometer: A two-thermometer device for simultaneously measuring

wet bulb and dry bulb temperatures of ambient air.

Range: Condenser water temperature increase (rise) in the condenser and

decrease (drop) in the cooling tower, 8F.Recirculation: Air leaving the cooling tower that re-enters the tower inlet,

elevating the tower entering wet bulb temperature.

Temperature (Air): Air temperature conditions that effect condenser water

systems include:

Dry Bulb Temperature: Dry bulb temperature (8F) of the air upwindfrom the cooling tower, unaffected by recirculation or other

cooling towers.

Ambient Wet Bulb Temperature: Web bulb temperature (8F) of the airupwind from the cooling tower, unaffected by recirculation or

other cooling towers.

Entering Wet Bulb Temperature: Wet bulb temperature (8F) of the airentering the tower, the product of the ambient air wet bulb

temperature and any recirculation air wet bulb temperature.

(Ideally, recirculation is zero and the entering wet bulb

temperature is the ambient wet bulb temperature.)


Wind Load: Structural load imposed by wind blowing on the tower casing,

psf.

9.4. COOLING TOWER HEAT TRANSFER

Heat is transferred from a water droplet to the surrounding air by both sensible

and latent heat transfer processes. Figure 9.2 illustrates a typical water droplet

and the heat transfer mechanisms.

This heat transfer process can generally be modeled using the Merkel

Equation [Eq. (9.2)]:

KaV=L ¼Z T1

T2

dT=ðhw 2 haÞ ð9:2Þ

where KaV/L ¼ cooling tower characteristic; K ¼ mass transfer coefficient,

lb water/hr ft2; a ¼ contact area/tower volume, 1/ft; V ¼ active cooling

volume/plan area, ft, L ¼ water mass flow rate, lb/hr ft2; T1 ¼ entering (hot)

water temperature, 8F; T2 ¼ leaving (cold) water temperature, 8F; T ¼ bulk water

temperature, 8F, hw ¼ enthalpy of air–water vapor mixture at bulk water

temperature, Btu/lb of dry air; ha ¼ enthalpy of air–water vapor mixture at wet

bulb temperature, Btu/lb of dry air.

The amount of heat removed from the water must be equal to the heat

absorbed by the surrounding air as shown by Eq. (9.3):

L=G ¼ ðh2 2 h1Þ=ðT1 2 T2Þ ð9:3Þ

FIGURE 9.2. Water droplet with surface film.

Chapter 9120

where L/G ¼ water to air mass flow ratio, lb of water/lb of air, h1 ¼ enthalpy of

air–water vapor mixture at inlet wet bulb temperature, Btu/lb of dry air,

h2 ¼ enthalpy of air–water vapor mixture at exhaust wet bulb temperature,

Btu/lb of dry air.

While the tower characteristic can be calculated from Eq. (9.2) using

numerical methods, it can also be represented graphically as shown in Figure 9.3.

In this figure, the graphical variables are defined as follows:

C0 entering air enthalpy at entering air wet bulb temperature Twb, Btu/lb

of dry air

BC initial enthalpy driving force

CD air operating line with slope L/G

DEF projecting the leaving air point onto the water operating line and then

onto the temperature axis yields the outlet wet bulb temperature, 8FABCD the area within this region is the graphical solution to the cooling tower

characteristic

For any given graphical solution, if the tower characteristic remains constant, an

increase in imposed heat rejection load will have the following effects:

1. The length of the line CD will increase and shift to the right.

2. Both T1 and T2 will increase.

3. The range and the approach areas will increase.

The Cooling Technology Institute (CTI) developed their CTI Blue Book many

years ago to provide a compendium of graphical solutions to Eqs. (9.2) and (9.3)

for use by cooling tower designers and manufacturers. In recent years, computer

FIGURE 9.3. Graphical representation of the cooling tower characteristic.


software to aid tower design has become common and numerical methods have

supplemented the CTI charts.

In HVAC applications, there are essentially no “custom” cooling towers

(except in a few very special cases). Manufacturers have designed a range of

cooling towers whose characteristics are well defined. For any HVAC

application, it is only necessary to use the specified flow rate, range, and

ambient wet bulb temperature to compute a required tower characteristic and then

test that required characteristic against those available in the product line.

Manufacturers provide catalog data and selection software for the purpose of

selecting a cooling tower for a specific application.But, since most HVAC designers

and the purchasers of towers are not expert in the use of a particularmanufacturer’s

catalog or selection software, it is always prudent to have the tower manufacturer

perform any required tower selection. Then, if an error is made, the manufacturer

will have to make good on the tower, not the designer or buyer.

The cooling tower’s evaporative cooling process can be shown on a

psychrometric chart as illustrated by Figure 9.4. The change in condition of a

point of dry air as it moves through the tower and contacts a pound of water (e.g.,

L/G ¼ 1) is shown by the solid line. Ambient air, at 788F DB and 50% RH, enters

the tower at Point 1 and begins to absorb moisture in an effort to gain equilibrium

with the water. This process continues until the air exits the tower at Point 2.

During this process, the following conditions changed:

1. The air enthalpy increased from 30.1 to 45.1 Btu/lb of dry air, an

increase of 15 Btu/lb of dry air. This heat came from the one pound of

water, reducing its temperature by 158F.2. The specific humidity of the 1 lb of air increased from 0.0103 lb of

moisture to 0.0233 lb of moisture. This increase of 0.013 lb of moisture

represents the amount of condenser water that was evaporated. Using

the latent heat of vaporization at 858F, the heat transfer due to

evaporation is 0.013 lb £ 1045Btu/lb or 13.6 Btu, 91% of the total

15 Btu of heat transferred to the air.

3. The condenser water temperature was reduced by 158F, while the air

temperature was increased by only 3.38F (from 78 to 81.38F).

9.5. COOLING TOWER PERFORMANCE FACTORS

The selection of a cooling tower for a specific set of required performance

parameters (condenser water flow rate, selected range, and ambient wet bulb

temperature) results in establishing a required cooling tower characteristic. But,

what is the impact if the required performance changes?

As shown in Figure 9.3, there are three factors that define the requirements

for a specific cooling tower characteristic:

Chapter 9122

FIGURE 9.4. Air–water temperature curve. (Courtesy of the Marley Cooling

Tower Company, Overland Park, Kansas.)


1. Entering (ambient) air wet bulb temperature.

2. Condenser water flow rate.

3. Approach.

Once a tower is selected, i.e., the cooling tower characteristic is established,

changes in any of these three factors may necessitate a change in the cooling

tower. Typically, the performance parameters that are subject to change include

the following:

1. Increase of entering wet bulb temperature. Too often, the wrong

ambient wet bulb temperature is selected for tower sizing.

2. Increase of rejected heat load. This may dictate increasing the

condenser water flow rate and/or increasing the range.

A change in wet bulb temperature and/or a change in range will result in a change

in approach. Therefore, there are three performance impact relationships as

shown in Figures 9.5–9.7.

Figure 9.5 illustrates the impact of changing the tower approach by either

changing the design wet bulb temperature and/or changing the condenser water

range. Obviously, a relatively small increase in required approach will require a

much higher cooling tower characteristic.

Figure 9.6 illustrates the impact of changing the tower load by changing the

water flow rate, while Figure 9.7 illustrates the impact of changing tower load by

changing condenser water range (without impacting approach). These two

variables have a smaller impact on the cooling tower characteristic.

FIGURE 9.5. Variation in tower size factor with approach.

Chapter 9124

9.6. BASIC COOLING TOWER CONFIGURATION

Chapter 11 describes the various configurations of HVAC cooling towers and the

particular advantages and disadvantages associated with each. However, basic

tower configurations are very simple, as shown in Figure 9.8, and are dictated by

(1) the direction of the air vs. water flow through the tower fill and (2) the location

of the tower fan(s)

FIGURE 9.6. Variation in tower size factor with condenser water flow rate.

FIGURE 9.7. Variation in tower size factor with range.


FIGURE 9.8. Cooling tower configurations.

Chapter 9126

Air/Water Flow: In counterflow towers, the water and air flow in opposite

directions, the water flows vertically downward and the air flows

vertically upward. In crossflow towers, the two flow streams are arranged

at 908 to each other, i.e., the water flows vertically downward through the

fill, while the air flows horizontally through it.

Fan Location: The tower is defined as forced draft when its fan(s) is

arranged to blow air through the tower. In this configuration, the fan(s) is

located on the entering-air side of the tower and the fill is under positive

pressure. In induced draft towers, the fan(s) is located on the leaving-air

side of the tower and the fill is under negative pressure.

For many years, one manufacturer (Baltimore AirCoil, Inc.) offered an

“induction draft” or “venturi” cooling tower, as shown in Figure 9.9, for the

HVAC market. In lieu of a fan, the return condenser water was sprayed

horizontally through high velocity nozzles into a venturi chamber, much like the

throat of an automobile carburetor. The high velocity water flow through the

venturi created a low pressure region that (1) induced airflow through the tower

and (2) tended to vaporize the water. Since air- and water-flow were in the same

FIGURE 9.9. Induction draft or Venturi cooling tower configuration.


direction, this tower would be classed as a “parallel” flow type. Since there is no

fill, this was a “spray fill” tower. The high water pressure required by the tower

signicantly increased the pump horsepower requirement, more or less offsetting

the fan energy savings. This problem was mitigated by using a two-speed pump

motor and operating the tower at low speed when the outside ambient wet bulb

temperature was low. This tower suffered from both performance and

maintenance problems and is no longer available except by special order.

Chapter 9128

10

Cooling Tower Components

As described inChap. 9, everyHVACcooling tower has six functional components:

fill, wet deck, basin, fan(s), inlet louvers, and drift eliminators. To this list we must

also add the two structural elements: the structural frame and the casing.

10.1. FILL

The function of the tower fill is to provide a large “contact area” between the

water flow and the air flow to promote evaporation and heat transfer.

10.1.1. Spray Fill

Very early, it was recognized that by breaking the water flow into droplets the

contact area between the water flow and air flow increased due to the increased

water surface area exposed to the air. Thus, the use of water sprays represents the

earliest concept used to improve the efficiency of evaporation and heat transfer.

In the mid-1800s, spray ponds were used to provide condensing for the ice-

making refrigeration systems then in use. These ponds, however, had relatively

low efficiency and high drift losses that motivated a search for a better method.

Then, in 1898, George Stocker is credited with building the first “cooling tower.”

This “atmospheric” tower consisted simply of wooden, louvered walls built

around a spray pond.

129

Over time, improvements such as draft chimneys were implemented and, in

the 1920s, the use of mechanical draft fans significantly improved airflow rates,

decreasing L/G ratios and improving performance.

Spray fill really isn’t “fill” at all. The water is sprayed through nozzles to

create as fine (small) droplet size as possible. The spray, then, is contained by the

tower casing (walls), creating a water-filled plenum through which the airflow is

directed. Spray density ranges from 3 to 7 gpm/cf of tower volume.

The overall efficiency of spray fill is relatively low, requiring large towers

with high airflow rates compared to other types of fill. Thus, spray fill is not used

in HVAC cooling towers.

10.1.2. Splash Fill

The next step in improving cooling tower efficiency was made by introducing

splash fill in the 1940s. Since most cooling towers of that era were constructed of

wood, splash fill was fabricated from wooden slats that were then stacked in the

tower air plenum to create a “cascading” path for the water as it fell by gravity

through the tower. Fill slats were either flat or triangular shaped, which tended to

reduce the airflow pressure drop, as shown in Figure 10.1.

Wooden splash fill remained in use until the 1980s, but modern splash fill is

typically constructed of vacuum-formed PVC slats or V-shaped bars that are

designed to provide good droplet formation and eliminate the rot problem

common to wood in wet environments.

Ceramic fill is an alternative splash fill material sometimes used in HVAC

applications. Formed from fired, ceramic shapes, this fill is stacked to create a

splash fill form. Ceramic fill is virtually impervious to rot or corrosion, but is very

expensive compared to other types of fill.

10.1.3. Film Fill

Film fill was introduced in the 1960s and created a new concept in cooling tower

design. Instead of breaking the water flow into fine droplets, film fill provides

large surface areas over which the water flows. With the large areas and slower

“falling” speeds than splash fill, the water forms a thin sheet or film as it flows

across the fill surfaces, creating large air–water contact area and efficient

evaporation heat transfer.

Film fill is typically constructed of thin sheets of vacuum-formed PVC,

stacked in vertical layers and formed with corrugations to ensure even water

distribution over the entire fill surface. This configuration has low resistance to

airflow; the air pressure drop through the tower is low. Typical film fill is shown

in Figure 10.2.

Chapter 10130

Film fill is both less costly and more efficient than splash fill. The higher

efficiency results in a significantly smaller tower (higher tower characteristic) for

an equivalent capacity. Thus, most HVAC towers applied today utilize film fill.

The disadvantage of film fill is that the large surface area and correspondingly

small water passages are sometimes susceptible to (1) clogging by debris build-

up (from leaves, dirty water, etc.) and (2) microbiological fouling, since

FIGURE 10.1. (a) Flat slat splash fill. (b) Triangular slat splash fill.

Cooling Tower Components 131

mechanical cleaning cannot be done unless the fill is removed from the tower and

disassembled (an expensive and time-consuming job).

10.2. STRUCTURAL FRAME

A structural framing system is required to support the wet deck, fan(s), fill, intake

louvers, and drift eliminators. Casing is required to enclose the fill and to create

the tower air and water flow path.

10.2.1. Wooden Structure

Early cooling towers were constructed with wooden structural frames and

redwood, because of its natural resistance to rot under wet conditions, was the

wood of choice. However, by the 1960s, the increased cost and reduced

availability of redwood forced the industry to shift to West Coast Douglas fir.

Fir, however, has almost no natural resistance to wet rot and must be

chemically treated for use in cooling towers. The use of wood preservatives has

significant undesirable environmental side effects, so, today, only under very

special circumstances are wooden towers considered for HVAC applications,

though they remain in common use for large industrial applications. If fir is to be

used, it should be specified to meet the requirements of CTI Standard 114 and all

fasteners should meet the requirements of CTI Standard 119.

One of the drawbacks to a wooden structure cooling tower is that most

property insurance companies will require that it be protected by a fire

sprinkler system installed in accordance with the National Fire Protection

Association (NFPA) Standard 214 (see Chap. 14). While it may seem odd that

a cooling tower needs fire protection, wood is highly combustible and fires in

FIGURE 10.2. Typical cooling tower film fill. (Courtesy of Evapco, Inc.,

Westminster, Maryland.)

Chapter 10132

cooling towers that are out of service has been a relatively common

occurrence. For this reason, and the fact that wooden towers in smaller sizes

are no longer cost competitive, wooden structure cooling towers are rarely

used for HVAC applications today.

10.2.2. Steel Structure

Since the 1950s, steel has become the structural framing system of choice for

HVAC cooling towers. To reduce the effects of corrosion in a wet tower

environment, two types of steel are used, as follows.

10.2.2.1. Galvanized Steel

Mild carbon steel is coated with zinc, lead, and aluminum in a process called

“galvanizing” to provide a corrosion barrier between the underlying steel and the

air/water on the surface.

In the galvanizing process, the steel to be coated is first cleaned in a

caustic bath followed by pickling in a dilute hydrochloric or sulfuric acid bath.

The steel is then subjected to a fluxing operation with a zinc ammonium

chloride solution.

Early galvanizing used the “wet kettle” method, in which a floating layer of

zinc ammonium chloride flux was placed on the surface of molten zinc in a large

open kettle. A section of sheet steel was then immersed in the molten zinc,

passing through the layer of flux and remaining in the molten bath until the

temperature of the steel was the same as the molten zinc. The zinc bath was then

swept free of the remaining flux and the sheet removed. The sheet was then

immersed in a water quench bath to cool it for handling. The quench water was

usually treated with sodium dichromate to provide further protection.

Since the 1980s, galvanizing has typically been done by the “dry kettle”

method. The steel sheet, now in continuous rolls, is first immersed in a kettle of

zinc ammonium chloride flux and then removed and allowed to dry. The fluxed

steel sheet is then dipped into the molten zinc bath. It is removed after heating and

sent through a water quench bath.

While the dry kettle method has been found to provide a higher quality

surface finish, using a much cleaner process, there remains a significant argument

in the industry relative to the quality of the resulting galvanizing (see Sec.

13.2.2). Therefore, most cooling tower manufacturers and HVAC designers

recommend that a protective coating of epoxy or polymer finish be added to all

galvanized structural members.

One important note: sometimes literature will mention “hot dip galvanizing

after fabrication” as a method of ensuring good corrosion protection. However,

hot dip galvanizing after fabrication has not been available from U.S. tower

manufacturers since the late 1970s.


The thickness of the galvanizing film is rated in terms of “ounces of zinc

per square foot of metal surface.” Typical cooling tower galvanizing is

2.35 oz/ft2, but galvanizing as thick as 7.0 oz/ft2 is available from some

manufacturers.

Galvanized steel structural frames are typically assembled with galvanized

fasteners, although some tower manufacturers will provide an option for stainless

steel. Galvanized steel is never welded since welding (1) destroys the galvanizing

and (2) establishes a point for differential corrosion to start (see Chap. 13).

10.2.2.2. Stainless Steel

Stainless steels generally have less structural strength than mild carbon steel, but

are far more resistant to corrosion in wet environments. Stainless steel consists of

steel alloys (carbon steel plus molybdenum, nickel, and/or nitrogen) and

chromium. The corrosion resistance credited to stainless steels results from an

extremely thin layer of chromium oxide that forms on the steel surface when

exposed to air. Over time, if the layer is damaged, the underlying chromium

(minimum 10.5% of the alloy) is re-exposed and the oxide layer reforms.

Almost any steel cooling tower can be fabricated using stainless steel and

most tower manufacturers offer this option.

One advantage of stainless steel members is that they can be TIG or MIG

welded and the corrosion integrity maintained. Where structural fasteners are

required, they are also stainless steel. (Note, however, that stainless nuts and bolts

tend to “creep” with stress and age, so are difficult to remove and often cannot be

reused.)

10.2.3. Other Structural Systems

In the never-ending search to improve HVAC cooling tower corrosion resistance

and extend tower life, other materials are used for structural systems, as follows.

10.2.3.1. Concrete

Concrete has been used as combined structure and casing for large hyperbolic

utility cooling towers for many years. However, the material is not widely used in

HVAC applications due to its relatively high cost. One manufacturer does offer a

line of factory-fabricated, field-erected concrete towers, but they have not had

large sales.

Concrete, however, can become a desirable material for HVAC cooling

towers when it is necessary to “blend” the tower into a larger architectural

scheme. Using concrete, the tower can be constructed to almost any configuration

to “hide” its utilitarian function. (A prime example is the Dallas-Ft. Worth

Airport.)

Chapter 10134

10.2.3.2. Fiberglass

Fiberglass structural tower frames, joined with stainless steel fasteners, are

offered by a few tower manufacturers for smaller HVAC cooling towers. These

towers utilize structural framing members constructed of 1/4–1/2 in. thick

fiberglass.

10.2.3.3. Stressed Skin Fiberglass/Stainless Steel

In the 1980s, the “bottle” cooling tower, developed around the Pacific rim, came

to the United States. These towers were designed for field-assembly of two sets of

curved fiberglass panels, much like “pie slices,” that were bolted together atop a

simple, tubular stainless steel space frame. One set of panels formed the basin of

the tower, while the other set formed the enclosure/air plenum. Once assembled,

the panels acted together to form a uniform stressed skin structural system to

support the fill, the wet deck, and the fan. Unfortunately, these towers suffered

from severe performance and reliability problems and have generally been

replaced.

10.3. CASING

A cooling tower’s casing performs two roles. First, it forms an enclosure around

the fill to create a contained air path or plenum, forcing the airflow through the

fill. Second, it simply helps to keep the water inside the tower.

Wood-framed cooling towers originally used cedar or redwood planks as

tower casing. However, to reduce expense, cement-asbestos panels came into

wide use in the 1950s, only to be replaced by fiberglass or UV-inhibited plastic

panels in the 1980s as the environmental problems with asbestos were identified.

Galvanized steel-framed cooling towers are typically encased by

galvanized steel, fiberglass, or UV-inhibited plastic panels. (If galvanized steel

is used for the casing, it too should be coated with an epoxy or polymer final

finish.) Stainless steel- and fiberglass-framed cooling towers are normally

encased by stainless steel, fiberglass, or UV-inhibited plastic panels.

10.4. WET DECKS/WATER DISTRIBUTION

The wet deck is located at the top of the tower and its job is to distribute the

incoming warm condenser water as evenly as possible over the fill to ensure

uniform heat transfer.

In crossflow towers, the wet deck is not in the air stream and normally

consists of shallow basins or reservoirs with evenly distributed bottom “holes”

consisting of plastic, metering orifices. Warm condenser water is dumped into the

basin from the return piping so as to maintain a fairly uniform depth (4–6 in.).


Thus, the gravity water flow rate through each bottom outlet will be uniform and

the flow rate over each section of fill will also be the same.

Since the wet deck on a crossflow tower is open to the atmosphere,

removable basin covers must be installed to keep leaves and other wind-blown

debris out of them. The “wet deck” in counterflow towers is in the air stream. The

wet deck consists of spray nozzles that are used to both evenly distribute the water

flow over the fill and to provide initial water atomization (i.e., act as spray fill).

10.5. BASINS

The basin is located at the bottom of the tower and its job is to collect the cold

condenser water for supply to the condenser water pump. Most commonly, basins

are provided as an integral part of the cooling tower and are typically 12–18 in.

deep. Operating water depths are usually less than 12 in.

In cold climates, where tower freezing is a problem, or when it is necessary

to have a basin with more water volume capacity than normal, a remote basin

(sometimes called a “sump”) may be constructed, usually of concrete. Separate

sumps are sometimes used to allow the use of vertical turbine pumps (see

Chap. 11).

There are three critical aspects relative to the basin:

1. It must have sufficient capacity to contain the water that will flow to it

when the condenser water pump shuts down.

2. The volume of water stored in the basin must be great enough to fill all

empty piping upon start-up. Otherwise, the basin may be pumped dry.

3. There must be a sufficient height of water above the condenser water

supply connection to prevent vortexing and the introduction of air into

the water flow.

See Chap. 11 for further discussion of these aspects.

10.6. INTAKE LOUVERSANDDRIFT ELIMINATORS

Intake louvers are provided on all crossflow cooling towers to help control the

airflow over the fill. The louvers are spaced and slanted to direct air evenly into

the fill pack.

Drift eliminators are installed on the leaving side of the fill designed to

“trap” entrained water droplets and prevent them from leaving the cooling tower.

Fabricated with PVC or steel, these devices force the air to make three or four

high velocity directional changes (about 908 each). Water droplets, then, will

impinge on the eliminators and drain back into the fill. Typically, drift

eliminators are designed to limit the drift losses of an HVAC cooling tower to

0.1–0.25% of the condenser water flow rate.

Chapter 10136

10.7. FANS, MOTORS, AND DRIVES

One or more fans, connected to one or more motors via a drive assembly, provide

the motive power for airflow through a mechanical draft HVAC cooling tower.

10.7.1. Fans

Two types of fans are used; centrifugal fans and axial propeller fans. Both types

may be applied forced draft towers, but only propeller fans are used with induced

draft towers. Both types produce airflow through the tower by increasing the

static pressure (SP) and kinetic energy of the air.

With the centrifugal fan, air enters through the center portion of an impeller

wheel and exits through a scroll and outlet at right angle to the inlet paths. Axial

flow fans, by contrast, have a straight-through airflow path. Figure 10.3 illustrates

these two fan configurations.

Fan performance characteristics are plotted for a particular rotational speed

in revolutions per minute (RPM ), with airflow rate as the independent variable

and power, SP, and efficiency as dependent variables. Figure 10.4 is a typical

centrifugal fan curve. Since the fan curve is plotted for one fixed fan speed,

airflow changes are affected only by changes in the system resistance.

For fans with backward-inclined blades, the mechanical efficiency curve

peaks at a higher efficiency and at higher fractions of maximum airflow rates than

do fans with forward-curved blades. In addition, the horsepower requirement

FIGURE 10.3. (a) Typical centrifugal fan. (b) Typical axial propeller fan.


continues to increase through full load power as the flow rate is increased in a

forward-curved fan, whereas in the backward-inclined fan the horsepower

requirement peaks at between 60 and 80% of maximum airflow capacity.

Therefore, a forward-curved fan can overload its motor if allowed to operate

without restriction, while the backward-curved fan’s power requirement is self-

limiting.

Unfortunately, most centrifugal fans used in cooling towers are the

forward-curved type since (1) the fans are relatively small and (2) forward-curved

fans are far less expensive to manufacture.

In Figure 10.4, the shaded area represents the normal selection range

for a centrifugal fan, ideally at the highest point of the fan efficiency to

minimize fan losses and horsepower requirement for the given airflow

conditions.

Relationships between a fan and the system it serves are illustrated in

Figures 10.5–10.7. First, consider the SP required to force a certain volumetric

flow of air through a cooling tower. As the volumetric flow rate increases, the SP

FIGURE 10.4. Typical performance characteristics of fans: (1) CFM/Static

pressure curve; (2) BHP curve; (3) Mechanical efficiency curve.

Chapter 10138

FIGURE 10.5. (a) Typical cooling tower system curve: airflow vs. static pressure.

(b) Typical cooling tower fan performance curve: airflow vs. static pressure.


FIGURE 10.6. (a) Change in tower system curve with increased resistance,

curve A to curve B. (b) Change in fan performance curve with increased speed,

curve 1 to curve 2.

Chapter 10140

required also increases, but increases approximately as the square of the

increased flow.

Figure 10.6 illustrates changes in the cooling tower resistance and fan

characteristics. Any time the resistance changes (due to dirty intake louvers,

clogged fill, etc.), a new system characteristic curve is generated. Similarly, if the

fan speed is changed, a new fan curve is generated. When a particular fan

operating at a particular speed is installed in a particular cooling tower, the only

stable operating point is the one that satisfies both the fan and the system

characteristic curves, i.e., at the intersection of the two curves.

Approximate relationships for assessing the effects of changes in fan speed,

airflow rate, and fan power is knows as the fan laws:

1. Volumetric airflow rate (CFM) is directly proportional to fan speed

(RPM).

2. Static pressure is proportional to the square of the fan speed.

3. Brake horsepower (BHP) is proportional to the cube of the fan

speed.

FIGURE 10.7. Fan and tower curves imposed to show determination of

operating point.


The mathematical expressions for the fan laws for changing conditions from

state “1” and state “2” are shown by Eqs. (10.1–10.3).

RPM2 ¼ RPM1ðCFM2=CFM1Þ ð10:1Þ

SP2 ¼ SP1ðCFM2=CFM1Þ2 ð10:2Þ

BHP2 ¼ BHP1ðCFM2=CFM1Þ3 ð10:3ÞAt any airflow and pressure condition, the theoretical fan BHP can be computed

utilizing Eq. (10.4).

BHP ¼ CFM{0:0003þ ½SP=ð6346 EFFÞ�} ð10:4Þwhere BHP is the total theoretical brake horsepower; CFM, the total airflow

through the tower; SP, the static pressure drop through the tower, wg; EFF, the

fan efficiency: backward-inclined centrifugal 0.70, forward-curved centrifugal

0.60, propeller 0.70.

Cooling towers may be equipped with a single fan or with multiple fans. If

multiple fans are provided, the tower design must be such that it can continue to

operate when one or more fans are off. The most common approach to this

requirement is to provide internal baffles to form individual fan plenums and

prevent air discharged by one fan from entering the discharge plenum of an

adjacent fan. Another approach is to use backdraft dampers on each fan discharge

to prevent reverse airflow through the fan when it is off.

10.7.2. Motors

There are three basic types of motors used in HVAC applications. They are as

follows.

1. Open Drip Proof (ODP ): ODP motors are constructed with openings in

the motor casing that provide a cooling air path to the motor windings.

ODPmotors are designed for dry applications and, if installed outdoors,

must be weather-protected.

2. Totally Enclosed Fan-Cooled (TEFC ): The TEFC motor shaft extends

from the motor casing to drive a cooling fan installed within a shroud

on the nondrive end of the motor.

3. Totally Enclosed Air-Over (TEAO): The TEAO motor windings are

enclosed in a moisture-tight enclosure. The casing is provided with large

fins designed to radiate the motor heat to an air stream passing over them.

ODP motors are never used in cooling towers simply because they are not

designed for outdoor, wet locations (and will quickly fail in this environment).

TEFC motors are typically applied in cooling towers where the motor is located

outside the tower airflow path. TEAO motors are used where the motor is located

Chapter 10142

within the tower airflow path. Both TEFC and TEAO motors have the advantage

of better cooling than ODP motors and can generally be operated continuously at

up to 15% overload without failure, while an ODP motor quickly overheats and

fails under almost any overload condition.

Using an “energy efficient” motor in lieu of a “standard” motor will reduce

fan energy use by 4.6–6.8%, depending on the motor size, and given the small

price premium is an excellent choice. A “premium efficiency” motor can reduce

energy use by an additional 1–3%, depending on the motor size, but the cost

increase for these motors is rarely justified for cooling tower applications.

Because cooling tower fans must cycle off and on for capacity control

(see Chap. 12), it is important that tower motors be selected for frequent start/stops.

Table 10.1 provides motor ratings for intermittent duty. In Table 10.1, “A”

represents the maximum number of starts per hour. “B” is the product of the

maximum number of starts per hour times the motor loadWk2. “C” is the minimum

rest period or off time, in seconds, between starts. Thus, the maximum allowable

number of starts per hour is the lesser of “A” or “B” divided by the load Wk2.

To reduce the wear and tear of fan cycling capacity control, two-speed

motors can be used on belt-drive cooling towers. These motors are available in

either single winding or dual winding configuration.

TABLE 10.1. Approximate Motor

Allowable Starts and Starting Intervals

(NEMA Design A and B motors,

1800 rpm, 60Hz)

Motor HP A B C

1 30 6 38

1-1/2 26 9 38

2 23 11 39

3 20 17 40

5 16 27 42

7-1/2 14 39 44

10 12 51 46

15 11 75 50

20 10 99 55

25 9 122 58

30 8 144 60

40 7 189 65

50 7 232 72

60 6 275 85

75 6 338 90

100 5 441 110


Single winding motors utilize their entire winding whether on high speed or

low, but the number of active poles is reduced from 8 at low speed to 4 at high

speed. Since the motor speed is a function of the number of poles, low speed is

always exactly half of the high speed (typically 1800/900 rpm).

Dual winding motors have two sets of windings, one that is used on high

speed and the other on low speed. The advantage of this configuration is that the

ratio between high and low speed can be something other than 2:1.

As an alternative to a single-speed motor, the tower can be equipped with

two separate motors, one to operate at high speed and the second to operate at low

speed, again at a 2:1 speed ratio. The low-speed second motor, often called a

“pony” motor, will (based on the fan laws) be about 1/8th the horsepower of the

high-speed motor. During tower operation, when operating at high speed, the

pony motor will “free wheel.” Then, when switched to low speed, the pony motor

drives the fan(s) and the larger high-speed motor will free wheel. (The idling

motor does not act as a generator, simply because there is not excitation current

available to it. There are, however, additional motor windage and drive losses

that must be carried by the operating motor.)

When variable frequency speed drives (VFDs) are applied to cooling tower

capacity control (see below and Chap. 4), since the output of a VFD is not a true sine

wave, the fan motor must be rated for the potential of increased heat build-up.

10.7.3. Mechanical Drives

While small centrifugal fans may be designed to operate at motor speeds

(1750–1800 rpm) and be directly connected to a motor, most cooling tower fans

are designed to operate at slower speeds since the SP losses through towers are

relative low (1 in. wg or less) and high-speed fan operation is simply not

needed. Two types of mechanical drives are applied to HVAC cooling towers,

as follows.

10.7.3.1. Belt Drives

Belt drives, as shown in Figure 10.8, are used in a wide range of HVAC cooling

tower configurations. These drives perform two functions: they transfer the

motor power to the fans, while reducing the motor speed to the required fan

speed. The power transfer dictates the belt(s) load rating, while the sheaves (one

on the motor and one of the fan) dictate the speed change in direct ratio to their

diameters (a 5 in. motor sheave and a 10 in. fan sheave will reduce the speed

by half).

Historically, V-belt drives, with multiple belts selected to meet the required

load rating, have been applied. Efficiency of power transfer can be improved by

about 3% if “cogged” V-belts are utilized. However, because of higher

maintenance requirements due to alignment and stretch problems with any type

Chapter 10144

of V-belt and to improve power transfer even more, many cooling towers today

utilize the single “power” belt. This single belt is much wider than the V-belt for

the same load and may have teeth or ridges that interlock with its sheaves to

prevent slippage (and “squeal”) on start-up.

10.7.3.2. Gear Drives

Gear drives (or gear “boxes”), illustrated in Figure 10.9, are used by some

manufacturers on larger towers. Here, a shaft connects the motor to a gearbox that

performs three functions: it transfers drive power from the motor to the fan while

changing the rotational direction by 908 (from horizontal to vertical) and reducing

the motor speed to the required fan speed.

Gear drives are typically applied only to larger towers with 50 hp or larger

fan motors. These drives typically use spiral bevel and helical gear sets and,

depending on the speed reduction ration and the motor horsepower, may have a

single stage or multi-stage gear set.

The service life of a gear drive is directed related to the “surface durability”

or “wear hardness” of the gears and the number of hours the drive is operated.

The American Gear Manufacturers Association (AGMA) has published Standard

420 to establish requirements for geared speed reducers. Cooling tower gear

FIGURE 10.8. Typical cooling tower belt drive assembly. (Courtesy of the

Baltimore Aircoil Company, Baltimore, Maryland.)


drives should be selected in accordance with AGMA Standard 420. CTI has

published Standard 111 that recommends “service factors” specifically for

cooling towers that should be applied to gear drives.

High quality bearings are critical to gear drive life and bearings should be

selected on the basis of a 100,000-hr L-10 life. (L-10 life is defined as the life

expectancy, in hours, during which 90% or more of the bearings will still be in

service.) The incremental cost for this bearing life is small and, for HVAC

cooling towers, generally ensures that bearings will need replacement only every

15–25 years.

The gear drive lubrication system must be designed to provide reliable

lubrication in both forward and reverse operation of the drive.

FIGURE 10.9. Typical cooling tower gear drive assembly. (Courtesy of The

Marley Cooling Tower Company, Overland Park, Kansas.)

Chapter 10146

11

Tower Configuration andApplication

11.1. TYPES OF COOLING TOWERS

Almost all HVAC cooling towers are film or splash fill, mechanical draft type.

Basic HVAC tower configurations are dictated by (1) the direction of the air vs.

water flow through the tower fill and (2) the location of the tower fan(s), as shown

in Figure 9.9:

Air/Water Flow Arrangement: In counterflow towers, water and air flow in

opposite directions: the water flows vertically downward and the air flows

vertically upward. In crossflow towers, the two flow streams are oriented

90 degrees to each other: the water flows vertically downward while the air flows

horizontally.

Fan Location: The tower is called forced draft when its fan(s) are arranged

to blow air through the tower. In this configuration, the fan(s) are located at the air

intake(s) of the tower and the fill is under positive pressure. In induced draft

towers, the fan(s) is located at the air exit(s) of the tower and the fill is under

negative pressure.

Figures 11.1–11.4 illustrate four common configurations typical for

HVAC cooling towers: forced and induced draft counterflow and forced and

induced draft crossflow.

147

FIGURE 11.1. Typical forced draft crossflow cooling tower. (Courtesy of the


FIGURE 11.2. Typical induced draft crossflow cooling tower. (Courtesy of The

Marley Cooling Tower Company, Overland Park, Kansas.)

Chapter 11148

FIGURE 11.3. Typical forced draft counterflow cooling tower. (Courtesy of Tower

Tech, Inc., Oklahoma City, Oklahoma.)

FIGURE 11.4. Typical induced draft counterflow cooling tower. (Courtesy of

Evapco, Inc., Westminster, Maryland.)

Tower Configuration and Application 149

11.1.1. Counterflow vs. Crossflow

There are waterside and airside advantages and disadvantages to both counterflow

and crossflow cooling towers, as summarized in Table 11.1. However, the largest

differences between these types of cooling towers are as follows:

TABLE 11.1. Counterflow Vs. Crossflow Cooling Towers

Configuration Flow Advantages Disadvantages

Crossflow Waterside Lower pump head, pump

power requirement, and

pumping energy

Potential orifice clogging

and poor water

distribution over fill

Easier access to wet deck

or maintenance

Wet deck basin may house

biological fouling

Better acceptance of

variation in water

flow with economizer

Larger tower footprint

Airside Lower static pressure loss

with lower fan power

requirement and energy

consumption

Large inlet louver surface

area makes icing

control more difficult

Reduced drift

Reduced recirculation

Requires fewer cells for

larger capacities

Counterflow Waterside Spray distribution

improves water

droplet size

Increased pump head due

to spray nozzles, pump

power requirement, and

pumping energy

Tower is taller and

increased height

accommodates

closer approach

Spray nozzles are difficult

to access and clean

Airside Counter flow improves

heat transfer

Higher static pressure

losses, fan power

requirement, and

energy consumption

High inlet velocities

may pull debris

into basin

Tendency for uneven

airflow across fill,

reducing tower

efficiency

Chapter 11150

Size/Arrangement: For any given capacity, counterflow towers will

typically have a smaller footprint than crossflow towers, thus requiring less space

for their installation. However, this size advantage may disappear when multiple

tower cells are required to meet the imposed cooling load:

1. Small counterflow towers may require air intake on only one side and it

is relatively simple to arrange multiple towers in a line, spaced just far

enough apart for maintenance (about 4 ft).

2. For larger counterflow towers, two cells can be joined, which

effectively reduces the air entry condition for each cell from 4 sides to 3

sides. If more than two cells are required, the next 1 or 2 must be spaced

at least one tower width away to ensure proper air entry.

3. Multiple crossflow tower cells can be joined side-by-side to form a

tower as long as desired. However, it is better to limit this side-by-side

arrangement to only 3 or 4 cells and to provide an airflow and

maintenance separation of at least one tower width between each set of

cells. When more than four cells are joined, there are often adverse

airflow patterns at the middle cell(s).

Fan Horsepower/Energy: Counterflow cooling towers generally have

higher airflow static pressure losses than crossflow cooling towers, requiring

more fan horsepower and resulting in greater fan energy consumption. These

higher losses are due primarily to counterflow towers being taller, with a greater

fill height, and to the water spray wet deck.

Pump Horsepower/Energy: The nozzles used in the counterflow cooling

tower spray wet deck require, typically, 5 psig residual water pressure to provide

proper atomization. This, coupled with the fact that counterflow towers tend to be

taller than crossflow towers, results in increased pump head, pump motor

horsepower, and pump operating energy.

In general, crossflow towers are the better selection when it is desirable to

minimize tower fan energy consumption, minimize pump size and pumping

energy, and provide ease of maintenance. The crossflow tower is better with

waterside economizer applications.

The counterflow tower is the better selection with the available space

(footprint) is limited and/or where icing during winter operation is a concern. The

counterflow tower may also be the better selection when very close approach is

needed (less than 58F).

11.1.2. Mechanical Draft

As discussed in Chap. 9, almost all HVAC cooling towers utilize mechanical

draft since the tall draft chimney and large size required by natural draft towers

would be almost impossible to incorporate in most building designs.


Crossflow towers may utilize forced draft or induced draft. Single flow

forced draft crossflow towers (1-cell) utilizing propeller fans are available in

smaller capacities (from 30 to about 250 tons). The fan(s) is mounted vertically in

the tower air inlet and forces the air through the fill and out the exit on the

opposite side of the tower.

As cooling load increases, more heat transfer surface is needed and the

double flow crossflow tower, with up to about 1100 ton capacity in 1 factory-

assembled cell, is typically utilized since it is considerably less costly to use one

large cell than to use two smaller cells. In this configuration, it is more efficient to

use one large fan to induce airflow through the tower rather than multiple fans to

force the airflow. The induced draft fan is mounted at the top of the tower and

induces airflow from each end, through the fill, and into a common exit air

plenum.

Counterflow towers may also utilize forced draft or induced draft up to

about 1300 tons capacity in one factory-assembled cell. Both propeller and

centrifugal fans are used for forced draft, but only propeller fans are used for

induced draft.

In forced draft counterflow configurations with propeller fans, the fan(s) is

located at the bottom of the tower, which requires a very special basin design

since the air must flow through the basin. In these towers, a water collection

system, consisting of sloped channels above the fans, move the water from the

bottom of the fill to an adjacent enclosed basin or tank.

Induced draft counterflow towers have a far simpler configuration. The

fan is located at the top of the tower and the air intakes are located between

the top of the basin and the bottom of the fill, typically on all four sides of the

tower. As for crossflow towers, the largest single tower cell that is available is

about 1100 tons.

Generally, there is no advantage of forced draft over induced draft or vice

versa. However, in applications where cooling towers are used in the winter and

there is the potential for icing, forced draft towers, particularly those with

centrifugal fans, should be avoided. The forced draft fans are located in the

entering cold air stream rather than in the warm leaving air stream, making them

highly susceptible to icing and potential mechanical failure. Thus, for cold

weather operation, the induced draft tower is the better choice since both the

potential for icing is reduced and, as discussed in Sec. 14.1, measures to control

any icing that does occur are much easier to implement.

11.2. CAPACITY AND PERFORMANCEPARAMETERS

In Chap. 9, two factors were defined as establishing the HVAC cooling tower

performance requirements:

Chapter 11152

1. The cooling load imposed on the tower, which is defined by the

condenser water flow rate and the selected range.

2. The approach, which is dictated by the design outdoor wet bulb

temperature and the required condenser water supply temperature.

11.2.1. Temperature Range and Approach

As defined in Sec. 9.3, range is the temperature difference between the condenser

water entering (condenser water return) and leaving the tower (condenser water

supply), while approach is the difference between the condenser water supply

and the entering air wet bulb temperature. These temperature differences are

illustrated in Figure 11.5.

The vast majority of HVAC condenser water systems are designed for a

108F range. This range was established many years ago based on desired flow

velocities in condenser tubes and maintaining the approach at a reasonably high

level. The “standard” selection criteria for water chillers is 858F condenser water

supply temperature and a 108F range, resulting in a 958F condenser water return

temperature.

Since the design ambient wet bulb temperature is fixed, the approach is

changed by changing the range (and resulting flow rate) and/or changing the

required condenser water supply temperature. From Figures 9.5 through 9.7, it is

obvious that there is a trade-off between approach and range—increasing range

reduces the required tower characteristic, but if increased range reduces the

FIGURE 11.5. Definition of “range” and “approach” for condenser water

systems.


condenser water supply temperature, the resulting reduced approach quickly

increases the required tower characteristic.

If the range, and the resulting condenser water flow rate, is a design option,

it is best to evaluate a number of cooling towers, each at a different range from 8

to 148F, to determine the optimum selection. However, if the resulting approach

for any range is less than 58F, the size of the tower will increase exponentially forthe same capacity and, at 38F, no further reduction to approach is possible with

current cooling tower technology.

11.2.2. Ambient Wet Bulb Temperature

Appendix A lists the recommended design outdoor ambient wet bulb temperature

for selected locations in the United States. These values range from 75 to 818Fand, with the standard 858F condenser water supply temperature, approach will

vary from 4 to 108F (with 6–88F most common).

All cooling towers should be selected on the basis of the design ambient

wet bulb temperature. But, the actual entering wet bulb temperature may be

higher due to recirculation. Normally, there is always some recirculation, but it is

usually small enough to ignore, e.g., the ambient and entering wet bulb

temperatures are essentially the same. However, poor tower design or poor

placement can create excess recirculation and the entering wet bulb temperature

may be several degrees above the ambient level. Section 11.4 addresses measures

required to minimize this condition.

11.2.3. Condenser Water Heat Rejection

The total heat rejection to the condenser water system is determined from

Eq. (9.2) and Table 9.1. Typically, the range for an HVAC condenser water

system is selected to be 108F. At this range, the required condenser water flow

rate, in terms of “gpm per ton of cooling load” can be computed as shown in

Table 11.2.

TABLE 11.2. Required Condenser Water Flow Rate for 108F Range

Compressor

input power

(kW/ton)

Total heat rejection

(Btu/ton)

Fcdw

(gpm/ton of cooling load)

Fcdw Reduction

(%)

0.8 15,000 3.00 0

0.7 14,640 2.93 2.0

0.6 14,280 2.86 4.7

0.5 13,920 2.78 7.3

0.4 13,680 2.74 8.7

Chapter 11154

For many years, “standard” condenser water conditions were based on

chillers with a power input of 0.8 kW/ton. Older literature will sometimes refer

to a cooling tower ton as 15,000Btu/hr and current manufacturer ratings are

still based on 3.0 gpm/ton as the “standard” flow rate. None of these old data

are valid today. Chillers can be purchased with power input requirements as

low as 0.4 kW/ton, so the actual condenser water flow rate for any application

must be computed based on the actual compressor power input, as shown in

Table 11.2. This, in effect, minimizes pumping energy and the cooling tower

size.

11.3. CHILLER/COOLING TOWERCONFIGURATION

When there is only one chiller, the only approach is to select one cooling

tower to serve that chiller’s condenser. However, when there are multiple

chillers, there are three options for designing the condenser water system as

shown in Figure 11.6:

1. Each chiller can have an individual, dedicated cooling tower.

2. Multiple cooling towers or a large multicell tower can be connected to a

common condenser water system to serve all of the chillers. Each

chiller, however, must have a 2-position isolation control valve. The

advantage to this configuration is versatility—any tower and any pump

can serve any chiller.

3. Multiple cooling towers or large multicell towers can be connected to a

common condenser water system with an individual pump serving each

chiller. This configuration eliminates the need for chiller isolation

valves and is the recommended configuration when the chillers have

different capacities. (The downside of this configuration is, of course,

loss of the versatility provided by the common parallel pumping

arrangement.)

With individual, dedicated towers, individual condenser water pumps and

supply and return condenser water piping between each chiller and its tower

is required. The cost of this piping is typically higher than the cost for a

single (but larger) condenser water supply loop serving all chillers from all

towers.

The multicell approach has a number of operating advantages:

1. More than one cell can be used to serve a single chiller, providing a

lower condenser water supply temperatures and reducing chiller energy

consumption.


2. The failure of tower cell does not necessarily mean that the any specific

chiller cannot be utilized, since all chillers are served by all of the tower

cells.

3. An additional, redundant tower cell can be included to provide “back-

up” in the event of a tower cell failure or to maintain the required

condenser water supply temperature in periods of higher than expected

wet bulb temperature.

FIGURE 11.6. (a) Multiple chiller/tower configuration, option 1; (b) multiple

chiller/tower configuration, option 2; (c) multiple chiller/tower configuration,

option 3.

Chapter 11156

4. Only one tower make-up water control system is required, rather than

one for each tower under the first option, reducing maintenance

requirements.

5. One cell (or more) can be shutdown for maintenance without

interfering with a specific chiller’s operation.

A disadvantage to the multicell approach is that automatically controlled two-

position isolation valves are required on both the inlet and outlet of each tower

cell. These isolation valves eliminate water flow through cells that are not active.


(continued)


11.4. TOWER PLACEMENT AND INSTALLATION

The best place to locate any HVAC cooling tower is on the roof of the building it

serves. This placement satisfies two conditions that are critical to proper tower

performance:

1. The elevation of the tower’s operating water level must be above the

condenser water pump (and any piping between the pump suction and

the tower basin).

2. Towers must have sufficient clearance from walls and other

obstructions around them to prevent recirculation of discharge air

back into the inlet. If there are multiple towers, they must be located to

avoid the discharge of one tower entering any adjacent tower.


Chapter 11158

Generally, a building roof is clear of obstructions that can result in

recirculation, provided that obstruction by aesthetic screening is not

added.

Section 11.5 addresses condenser water pump placement and piping options. This

section addresses tower placement and installation to avoid recirculation of tower

airflow if the tower must be located on the ground.

The airflow requirement of an HVAC cooling tower is approximately 270–

300 cfm/ton, which is equivalent to 90–100 cfm/gpm. Thus, (1) sufficient free

area around the tower for the unobstructed flow of large air quantities must be

provided and (2) the tower must be located sufficiently clear of obstructions to

prevent discharge air from re-entering the tower (or entering an adjacent tower).

Ground level placement considerations for a tower depend, initially, on the

type of tower—crossflow or counterflow. Crossflow towers have an air intake on

either one (single flow towers) or two sides (double flow towers), while

counterflow towers have air intakes on all four sides. However, for both types of

towers, the following recommendations apply:

Adjacency: Building walls, screen walls, trees, etc. form obstructions for

cooling towers if the tower is located too close to them. While most tower

manufacturers provide information about layout for their cooling towers, most of

the recommended clearances are minimal and are not sufficient to always ensure

proper tower performance. Therefore, the recommended clearance from adjacent

obstructions is shown in Figure 11.7.

FIGURE 11.7. Recommended cooling tower siting parameters.


The cooling tower should be located at least 30 ft away from adjacent

building walls. As buildings get taller, the obstruction formed increases and, thus,

the tower setback should be increased to at least half the building height to

prevent recirculation.

Cooling towers are rarely pretty, thus, architects will often specify that

screening (wood, masonry, or metal) be installed around the tower. To prevent

tower performance problems due to screening, the following design criteria for

screening must be followed:

1. The screening must provide at least 50% open area on the air intake

side(s) of the cooling tower and the open area must be sufficient to

allow for 600 fpm or low airflow velocity through the screen.

2. The top of the screening must be no higher than the tower air discharge

(see “Elevation”).

3. Provide clearance between the screening and tower air intake side(s)

of the greater of 10 ft or the width of the tower. If there are multiple

towers or multiple tower cells within a screened enclosure, these

clearances must be increased by at least 50% (to 15 ft). Maintenance

clearance between the screening and the other (nonactive) sides of

the tower must be at least 4 ft to provide adequate passage and

maintenance space.

Elevation: Ideally, any HVAC cooling tower should be installed so that the

operating water level is higher than all of the elements of the condenser water

system. In every installation, the operating water level must be higher than the

condenser water pump and all of the condenser water supply piping between the

tower basin and the pump. Thus, the pump is maintained at a “flooded” condition

and cannot lose prime.

Where the tower is installed close to a building (i.e., closer than the half the

building height or 30 ft, whichever is greater), the fan discharge elevation must be

at or above the adjacent building roofline or the top of any architectural screening

that is provided. As shown in Figure 11.8, this can be accomplished by

(1) providing a discharge cowl or plenum on the tower and/or (2) providing a

structural support system to elevate the tower to the proper height.

An elevating structural support system or “grillage” can be easily

fabricated from structural steel elements, pipe columns, etc. The discharge cowl,

however, must be engineered to minimize static pressure loss and avoid the need

for a larger fan motor and increased fan energy consumption. The cowl must have

a discharge taper (called an evase discharge ) in order to reduce the air discharge

velocity efficiently and regain the available velocity pressure to offset friction-

based static pressure losses.

Table 11.3 tabulates the static pressure regain factor for a cowl of length L

(ft), a fan discharge diameterD1 (ft), and a final discharge diameterD2 (ft). There

Chapter 11160

FIGURE 11.8. (a) Tower elevation, option 1; (b) tower elevation, option 2.

TABLE 11.3. Discharge Cowl

Static Pressure Regain Factors

(CRF)

Static pressure regain

factor of D2/D1 ratio

L/D1 1.2:1 1.4:1 1.6:1

1.0:1 0.26 0.27 0.22

1.5:1 0.27 0.33 0.31

2.0:1 0.29 0.36 0.36

3.0:1 0.31 0.40 0.42

4.0:1 0.32 0.42 0.44

5.0:1 0.33 0.43 0.46

7.5:1 0.34 0.44 0.49


is little improvement to the static pressure regain factor for an L/D1 greater than

7.5:1 or D2/D1 greater than 1.6:1.

The amount of regained static pressure is computed from Eq. (11.2):

SP ¼ ðV=4005Þ2 £ CRF ð11:2Þwhere SP is the regained static pressure, “WG”; V, fan discharge velocity, fpm;

CRF, cowl regain factor from Table 11.3.

The amount of static pressure regain can be subtracted from the static

pressure loss through the cowl, which can be determined from any duct sizing

nomograph using the average cowl diameter [(D1 þ D2)/2], the average airflow

velocity through the cowl, and the cowl length. Designed correctly, the discharge

cowl should impose little or no additional static pressure for the tower fan to

overcome.

Prevailing Wind: Since counterflow towers typically have air intakes on all

four sides, tower orientation relative to the prevailing summer wind is not

important. However, always attempt to orient a crossflow cooling tower as shown

in Figure 11.9. Wind can influence the performance of a double inlet tower by

creating an unbalanced airflow through the two sides. If a crossflow tower must

FIGURE 11.9. Crossflow tower placement relative to the prevailing wind.

Chapter 11162

be installed with one inlet in the direction of the prevailing wind, the tower

manufacturer must be apprised of this condition so that the proper sizing

adjustments can be made.

Multiple Towers or Tower Cells: Figure 11.10 illustrates the recommended

layout for multiple towers or tower cells for both crossflow and counterflow

cooling towers. As discussed earlier, no more than 4 crossflow cooling tower cells

should be close-coupled. For counterflow towers, usually only 2 cells can be close-

coupled and even then the tower manufacturer must guarantee that the individual

cell performance will not be impaired since each cell is losing one intake side.

Again, wind direction is relatively unimportant to counterflow towers.

However, for crossflow towers, recirculation from one tower bank to an adjacent

one can be significantly increased by the prevailing wind. Figure 11.11 shows

FIGURE 11.10. (a) Recommended multicell crossflow cooling tower

arrangement (4 or more cells); (b) recommended multicell counterflow cooling

tower arrangement (2 or more cells).


recommended tower layouts relative to the prevailing winds for crossflow tower

bank installations.

Dunnage and Grillage: Dunnage is the steel and/or concrete structural

foundation upon which the tower support grillage rests. The tower grillage

FIGURE 11.11. (a) Recommended multicell crossflow cooling tower

arrangement relative to the prevailing wind, option 1; (b) recommended

multicell crossflow cooling tower arrangement relative to the prevailing wind,

option 2; (c) recommended multicell crossflow cooling tower arrangement

relative to the prevailing wind, option 3; (d) recommended multicell crossflow

cooling tower arrangement relative to the prevailing wind, option 4.

Chapter 11164

routinely consists of structural steel members arranged to support the tower at the

required bearing points.

On the roof of a building, dunnage should consist of structural column

extensions so that the tower weight is transferred directly to the building

foundation elements. Never attempt to install a tower directly on the roof

structural system! Roofs are never designed for the types of loads introduced by a

cooling tower and its water load. Even if the roof structure is designed to support

the operating tower weight, vibration produced by the tower will assuredly be

transferred to the roof structure and create serious noise problems. The columns



should be extended 3–4 ft above the roof to provide clearance for roofing

maintenance.

On the ground, dunnage many consist of concrete piers or columns—each

with properly designed footings and reinforcement.

Table 11.4 provides approximate operating and shipping weights for

typical cooling towers.

Care must be taken to ensure that dunnage/grillage height is sufficient to

satisfy the tower operating water level elevation requirement.

11.5. COOLING TOWER PIPING

Three separate piping systems are required for each cooling tower, as shown in

Figure 11.12 and described in the following subsections.

11.5.1. Condenser Water Piping

Condenser water return (CDWR) piping delivers warm water from the condenser

to the tower, while condenser water supply (CDWS) piping delivers cool water

from the tower to the condenser. A manual isolation valve should be provided at

TABLE 11.4. Approximate Cooling Tower Weight

Approximate operating weight (lbs)Nominal

capacity

(ton)

Forced draft

crossflowaInduced draft

crossflowaForced draft

counterflowbInduced draft

counterflowa

100 4,300 9,000 12,000 5,000

200 8,200 14,000 12,000 6,500

250 9,500 14,000 13,000 7,500

300 14,000 14,000 13,000 8,500

350 16,000 18,000 16,000 10,000

400 16,000 18,000 16,000 11,000

450 19,000 22,000 19,500 14,000

500 19,000 28,000 19,500 14,500

600 — 29,000 19,500 17,000

700 — 35,000 22,600 19,000

800 — 37,000 22,600 22,000

900 — 44,000 23,600 28,000

1000 — 44,000 23,600 28,000

1500 — 75,000 47,200 43,500

2000 — 90,000 47,200 63,000

a Shipping weight is approximately 30–35% of operating weight.b Shipping weight is approximately 55–65% of operating weight.

Chapter 11166

FIG

URE11.12.

Typical

coolingtower

pipinginstallationschem

atic.

both the supply and return connections to the tower to facilitate tower

maintenance. Electric heat tape (or other type of heat tracing) may be required to

protect the condenser water from freezing during off-periods in the winter (see

Chap. 14).

Black steel or PVC piping materials are normally used for condenser water

service. In the past, galvanized steel piping was often used because it was thought

that this piping was more resistant to corrosion and the higher cost for galvanized

over black steel was justified. However, since galvanized pipe cannot be welded

without destroying the galvanizing at each joint, screwed connections are

required. In the larger pipe sizes required for condenser water systems, increased

labor costs have made screwed joints prohibitively expensive and galvanized

piping is rarely used today.

The CDWR piping between the tower and the condensers should be

insulated with at least 1 in. thick glass fiber insulation to reduce heat gain by the

condenser water to the chiller. At normal condenser water temperatures, a vapor

barrier is not required, but, if a waterside economizer cycle is incorporated, a

vapor barrier does become necessary.

Vortexing is a term used to describe the whirlpool effect that sometimes

occurs at the CDWS connection to the tower basin, particularly with a vertical

discharge connection. Vortexing causes air to be entrained in the water and can,

in extreme cases, cause pump cavitation and result in mechanical damage. Most

tower manufacturers offer an option for a side outlet discharge sump that can be

added to the basin. This type of sump will minimize air entrainment and should

always be used.

11.5.2. Make-Up Water Piping

Make-up water is potable water from municipal water systems and/or wells that

adds water to the condenser water flow to offset the water losses due to

evaporation, drift, and blowdown. Again, a manual isolation valve is required at

the tower connection point for maintenance purposes. Copper tubing or PVC

piping is normally used for make-up water piping and all piping should be

insulated with at least 1-in. thick glass fiber insulation with an integral vapor

barrier.

The make-up piping terminates at a make-up water control valve, which

may be either mechanically or electronically actuated in response to the water

level in the basin. Electric heat tape (or other type of heat tracing) is normally

required to protect the make-up water from freezing during off-periods in the

winter (see Chap. 14).

Water losses from an HVAC cooling tower result from evaporation, drift,

and blowdown. Each of these must be determined to properly size the make-up

water system.

Chapter 11168

Evaporation Loss: The heat of vaporization of water is 1045Btu/lb at 858F.Therefore, to determine the evaporation rate required for a given heat rejection

(condenser water cooling) rate, it is necessary to know the flow rate, the range,

and the heat rejection required. In the general case (a flow rate of 1 gpm with a

range of 18F), the evaporation requirement can be computed as follows:

Flow: 1 gal=min ðequivalent to 500 lb=hrÞ

Range: 18F

Heat load: By rearranging Eq: ð9:1Þ;

Q ¼ 1 gal=min £ 8:34 lb=gal £ 60min=hr £ 1Btu=lb F £ 1 F

¼ 500Btu=hr

Heat of vaporization: Qv ¼ 1045Btu=lb

Evaporation rate: E ¼ 500Btu=hr = 1045Btu=lb ¼ 0:478 lb=hr

Thus, dividing the evaporation rate of 0.478 lb/hr by the flow rate of 500 lb/hr

yields an evaporation rate of essentially 0.1% of the condenser water flow rate per

8F of the range, or approximately 1% of the condenser water flow rate for the 108Frange typically used in HVAC applications.

Drift Loss: Drift loss data can be provided by the manufacturer for the

specific cooling tower selection being considered. However, drift loss by HVAC

cooling towers is typically within the range 0.1–0.2% and most engineers simply

use the higher value when evaluating make-up water requirements.

Blowdown Loss: Water treatment programs for deposition control routinely

require blowdown, the intentional wasting condenser water to increase the

amount of make-up water to maintain the ratio of dissolved solids in the

condenser water at a desirable level (see Chap. 13). The water treatment program

will establish the required number of cycles of concentration, which is the ratio of

the amount of dissolved solids in the condenser water to the amount in the make-

up water. Once the cycles is determined, the amount of required blowdown can

be computed via Eq. (11.3)

BD ¼ E=ðCycles2 1Þ ð11:3Þwhere BD is the blowdown flow, gpm; E, evaporation rate, gpm; Cycles, cycles

of concentration established by water treatment program.

Typically, the cycles of concentration will be between 5 and 10, and the

lower value, which requires greater blowdown to achieve, is often used to

establish make-up water flow rates.


Thus, the maximum required make-up water flow rate, per gpm of

condenser water flow, can be conservatively computed as follows:

Evaporation: (0.01 £ 1 gpm) 0.0100 gpm

Drift: (0.0020 £ 1 gpm) 0.0020 gpm

Blowdown: [0.01 gpm/(5 2 1)] 0.0025 gpm

TotalMake-Up: 0.0145 gpm

A 500-ton chiller with a rated condenser water flow rate of 3 gpm/ton would have

a condenser water flow rate of 1500 gpm and the maximum amount of make-up

water required would be (1500 gpm CDW £ 0:0145 gpm make-up/gpm CDW) or

22.75 gpm. For an “optimum” chiller with a condenser water flow rate of

2.8 gpm/ton, the make-up water requirement would be 21.23 gpm at the peak load

of 500 tons.

A water meter, of a type satisfactory to the local municipality, should be

installed in themake-upwater supply line so that the amount ofwater consumed by

the tower can be accurately determined. Most municipal water and sewer systems

will provide a credit against sewer costs for water lost through a cooling tower.

11.5.3. Drain and Overflow Piping

Every cooling tower must have drain and overflow piping connected to a sanitary

waste water system. It is illegal to drain chemically treated condenser water into

storm or site drainage systems.

The overflow portion of the system is simply an open drain at the same

level as, or slightly above, the operating water level in the basin. If excess water

enters the basin, it automatically flows into the overflow drain and is wasted. A

manual isolation valves controls the basin drain connection, which allows the

basin water to be emptied for periodic maintenance.

11.5.4. Multiple Towers or Cells Piping

When two or more cells are piped together to act a one cooling tower, each cell is

typically provided with the piping systems outlined above. However, there are

some modifications and additions that must be considered.

Common Condenser Water Piping System: If each cell is piped from a

common condenser water piping system, automatic isolation valves on both the

supply and return are required to control flow through the cell (see Chap. 12).

Without the isolation valves, part of the flow may go through inactive cells,

destroying the ability to control the CDWS temperature.

The elevation and inlet pressure requirements for multiple cells must be

considered if different types of cooling towers or tower cells (or similar towers by

different manufacturers) are to be utilized together. First, the basin operating

water level for each tower or cell must be at the same elevation. For crossflow

Chapter 11170

towers, the wet deck basins must have the same elevation or flow will tend to

favor the lower basin, which will routinely overflow while the higher basin tends

to run dry. To avoid this problem, it is necessary to install the supply piping to

each cell at the same elevation above the higher inlet and letting the water fall

into the individual inlets. To prevent siphoning, a vent, as shown in Figure 11.13,

is required for the lower cell and the downcomer pipe must be sized on the basis

of gravity flow. (Also as shown in Figure 11.13, piping to each cell should be kept

as symmetrical as possible to reduce flow imbalance resulting from different

pressure losses through different flow paths.)

Even when multiple gravity flow inlets are at the same elevation, equal flow

distribution is, for all practical purposes, impossible to achieve and there will

always be slight flow imbalance from cell to cell.

For counterflow towers, water spray connections should be at essentially

the same elevation and fitted with nozzles requiring the same water pressure in

order to equalize static head requirements for the condenser water pump.

Avoid mixing crossflow and counterflow tower cells on the same condenser

water piping system. The different elevations and inlet pressure requirements for

the two types of towers make it essentially impossible to balance flow between

them well enough to maintain satisfactory operation.

FIGURE11.13. Condenser water supply piping schematic for multiple towers or

cells with different inlet elevations.


Common Basin: Where the cells share a common basin, or where individual

cell basins are connected together, two considerations must be addressed:

1. A single make-up water valve and level control is required, not one for

each basin. With multiple make-ups to a single basin, the controllers

will “fight” each other and level control will be impaired.

2. Multiple, individual cell basins can be connected together to form a

single basin by installing manufacturer-fabricated basin connectors

between each basin. These connections create large openings between

each cell basin, effectively creating one common basin. An alternative

approach would be to operate the tower with “dry” individual cell

basins and drain all supply condenser water to a common remote sump

(see Sec. 14.1).

Where each cell has an individual basin, the basins can be connected together

with an equalizer line, as shown in Figure 11.14. No matter how closely matched

the piping is between cells, flow to each cell is never exactly balanced. Without

the equalizer line, one cell basin would probably overflow while another was

pumped dry. The equalizer line allows water to flow between tower basins and

compensate for the minor flow imbalances between the cells that will exist.

Based on experience, the equalizer line is typically sized to accommodate

15% of the flow rate of the largest cell. On this basis, Table 11.5 summarizes

recommended equalizer line sizes relative to tower cell flow rate.

FIGURE 11.14. Recommended multiple tower or cell isolation and equalizer

piping schematic.

Chapter 11172

With an equalizer line, the water operating level will be essentially the

same in each cell basin. Therefore, it is critical that each cell be installed so that

its recommended operating water level is at the same elevation as adjacent cells.

This may require cell height adjustment(s) if different cell styles or cells by

different manufacturers are piped together.

11.6. PUMP SELECTION, PLACEMENT, ANDPIPING

For HVAC condenser water systems, centrifugal pumps (as discussed in Sec. 3.3)

are the most commonly used type of pump and inline, end-suction, horizontal

split case, and vertical split case pumps can all be used for condenser water

service. However, for some applications, vertical turbine pumps are preferred.

Vertical turbine pumps, as shown in Figure 11.16, have two or more impellers

mounted in stages along a common shaft, with the discharge of each stage serving

as input to the next. Both the motor and pump shafts are arranged vertically. This

pump is designed with a piped discharge connection, but the intake is designed to

be simply submersed in a tank or pool of water (such as a cooling tower basin).

Part of the head that must be produced by a condenser water pump consists

of the piping friction losses in the condenser water system (including piping,

fittings, valves, etc. and the chiller condenser) and these losses can be computed

the same as for chilled water systems (see Sec. 3.3.2). However, since the

condenser water system is an open system, two additional pressure conditions

must be evaluated:

1. Static head is the height, in terms of feet of water column that the

condenser water must be lifted and is normally simply the elevation

difference between the operating water level in the basin and the wet

deck, as shown by Figure 11.15. It makes no difference if the tower is

TABLE 11.5. Multicell Tower Equalizer Line Size

Largest cell flow rate

(gpm)

Equalizer line size

(in.)

Up to 240 4

241–627 6

628–1167 8

1168–1925 10

1926–2820 12

2821–3465 14

2336–3850 2@10

3851–5640 2@12


on the roof of a tall building or sitting on the ground outside the chiller

room, the static head is represented only by the height of the “open”

portion of the condenser water system, which is always the tower itself.

2. Required residual pressure at the wet deck is the final component of the

condenser water head requirement. For crossflow towers, the gravitywet

deck system imposes no residual pressure requirement. But, for

counterflow towers, the required spray nozzle pressure (typically

2–5 psig)must be included. (The residual head requirement is computed

by multiply the psig pressure requirement by 2.31 ft of water/psig.)

A check valve must always be installed at the pump discharge. The check valve

prevents condenser water in the piping on the discharge side of the pump from

“draining back” through the pump and into the basin while seeking an

equilibrium level (Bernoulli’s Law) when the pump is not operating.

With base-mounted end-suction or split case pumps, the pump may be

located near the cooling tower or may be located remotely in a mechanical

equipment room. In both cases, two conditions must be satisfied:

1. The pump and all condenser water suction piping (piping between the

cooling tower basin and the pump inlet) must be installed below the

base tower basing operating water level. If any section of the suction

piping is installed above the basin water level, a portion of the water

FIGURE 11.15. Condenser water pressure loss elements at the cooling tower.

Chapter 11174

will drain back anytime the pump is shutdown. Unfortunately, the next

time the pump is started, there will be an air gap in the piping and there

will be no condenser water flow.

2. The available net positive suction head (NPSH) must be equal to or

greater than the pump’s required NPSH in order to prevent pump

cavitation.

Cavitation is a condition that results when the pressure at the pump impeller inlet

is below the vapor pressure of the water, resulting in “flash” vaporization and the

formation water vapor bubbles within the flow stream. As the pump impeller

increases the pressure, these bubbles collapse and the resulting noise is very

similar to rattling marbles in a can. The pump can be damaged by cavitation

(localized pitting on both the impeller and the casing) and, thus this condition

must be avoided.

Available NPSH is determined from (Eq. 11.5).

NPSH ¼ BP2 H 2 SH2 VP ð11:5Þwhere NPSH is the available net positive suction head, ft of water column; BP,

the barometric pressure, ft of water, absolute; H, the pressure loss due to friction

in the suction piping, ft of water; SH, the static head on the suction side of the

pump (“lift”), ft of water; VP, the water vapor pressure, ft of water, at water

temperature.

The BP and VP values for typical HVAC conditions are summarized in

Tables 11.6 and 11.7.

Vertical turbine pumps are always located at the tower since the suction

side of the pump must be installed in the basin or in a sump or basin extension

specifically constructed for the pump. Vertical turbine pumps are a good option

TABLE 11.6. Water Vapor Pressure (VP)

as a Function of Temperature

Temperature

(8F)Vapor pressure

(ft. of water, absolute)

40 0.28

45 0.35

50 0.42

60 0.59

70 0.89

80 1.17

85 1.38

90 1.61

100 2.19


when it is necessary to locate the pump at the tower due to one or both of the

following conditions that would result in an NPSH problem for a horizontal

pump:

1. Long suction piping lengths to a remote pump.

2. The tower elevation is significantly below the elevation of a remote

pump.

Figure 11.16 illustrates a typical vertical turbine pump installation.

Side outlet sumps on HVAC cooling tower basins are designed to minimize

vortexing and associated air entrainment in the supply condenser water. But, as

the water flows through the cooling tower, it becomes aerated and leaves the

tower with some entrained air. In the pump suction piping, i.e., piping between

the tower basin and the pump inlet, some of this air will tend to separate. At the

pump, if an eccentric reducer is installed in a flat on bottom configuration,

released air will be “trapped” in the top of the suction piping, reducing the

effective pipe area and impeding flow. Additionally, free air can enter the pump,

causing a noise similar to cavitation. This condition can be simply avoided by

using concentric reducers or installing an eccentric reducer in a flat on top

configuration.

11.7. EVAPORATIVE CONDENSERS ANDCOOLERS

Evaporative condensers and coolers are “first cousins” of the cooling tower, the

difference being the use of a “closed” coil in lieu of “open” fill.

The vast majority of direct expansion (DX) cooling systems use ambient air

as their heat sink for condensing the refrigerant during the vapor compression

TABLE 11.7. Barometric Pressure (BP)

as a Function of Elevation

Elevation

(ft)

Atmospheric pressure

(ft of water)

0 (sea level) 33.9

500 33.3

1000 32.8

1500 32.1

2000 31.5

3000 30.4

4000 29.2

5000 28.2

Chapter 11176

refrigeration cycle. For comfort cooling, the refrigerant condensing temperature

is typically 110–1408F, resulting in 75–1008F refrigerant lift or energy input.

With evaporative cooling, the condensing temperature can be reduced to 95–

1058F, saving significantly on the system energy requirement.

The evaporative condenser, as illustrated schematically in Figure 11.17,

replaces the tower fill with a refrigerant-to-air cooling coil. Water is introduced

through a spray type wet deck in a counterflow configuration to the airflow. By

wetting the coil surface and allowing evaporative cooling to occur, the effective

air temperature is reduced by 10–158F, which results in significantly improved

heat transfer and a lower condensing temperature compared to dry air cooling.

Evaporative condensers are usually designed for saturated condensing

temperatures between 95 and 1058F with an entering wet bulb temperature of

78–808F.

Water flow to an evaporative condenser is typically between 1.0 and

1.25 gpm/ton simply because their range is higher (usually 25–308F). The

evaporation rate, however, is the same as for a cooling tower, 0.1% per degree of

range, and water treatment is required (see Chap. 13).

FIGURE 11.16. Recommended installation for remote condenser water sump

with a vertical turbine pump.


The evaporative cooler is configured the same as the evaporative condenser

except that a liquid, rather than refrigerant vapor, flows through the coil. This

device is sometimes referred to as a closed-circuit cooler since the liquid is not

circulated through an “open” cooling tower.

Evaporative coolers are larger, more expensive, and require higher fan

energy input than a cooling tower sized for the same capacity, range, and

approach. However, they allow for the cooling of liquids other than water, such as

the water–glycol mixes that are often used in data processing cooling and water

source heat pump systems.

The use of both evaporative condensers and evaporative coolers is

somewhat limited in HVAC applications and a detailed discussion is beyond the

scope of this text.

FIGURE 11.17. Evaporative condenser schematic.

Chapter 11178

12

Cooling Tower Controls

HVAC cooling tower control falls into four areas: start-up control, capacity or

condenser water temperature control, make-up water control, and operating

safety control.

12.1. START/STOP CONTROL

Obviously, the first step in controlling the operation of a cooling tower is to

simply turn it on and off when needed. For a “dedicated” tower, starting and

stopping is controlled by an electrical “interlock” with the chiller, so that the

tower is allowed to operate anytime the chiller is in operation. This interlock

first starts the condenser water pump(s) to establish flow and, then, the actual

tower operation is controlled by the capacity control scheme applied (see Sec.

12.2). In a multitower or multicell common condenser water system, the tower

start/stop control is an integral part of the capacity control scheme for the

system.

Figure 12.1 illustrates a common hardwired dedicated chiller, condenser

water pump, and cooling tower interlock scheme.

For any condenser water system, either single tower or multitower/cell, the

initial system start-up can create a problem. If the system has been shutdown

overnight, as often happens in the spring and fall, the condenser water

temperature in the tower basin may be relatively cold (below about 658F). If this

179

cold water is circulated to the condenser, the chiller may shut down by its safety

operating controls due to “low condenser water temperature” or “low refrigerant

temperature.” In either case, the chiller will simply not start until the condenser

water temperature is increased to above 658F. To prevent this problem, a

FIGURE 12.1. Dedicated chiller and cooling tower interlock wiring schematic.

Chapter 12180

modulating three-way control valve should always be included with HVAC

cooling tower systems.

The control valve is normally open to the bypass (normally closed to the

tower). Thus, upon start-up, all condenser water flow bypasses the tower and

simply circulates through the condenser. As the chiller loads, the condenser

begins to reject heat to the water, raising its temperature. As the temperature

rises, to setpoint (usually 70–758F), the control valve modulates the bypass port

closed, diverting the water to the tower, upon which the tower capacity control

scheme takes over.

The bypass control valve should be placed as close to the condenser as

possible, but it must also be at an elevation at or below the tower basin operating

water level if the bypass line is connected directly to the condenser water supply

line. If not, when the chiller and its condenser water pump shut down at night, the

bypass port reopens and the water in the bypass line will drain to the same level as

the tower operating water level. This, in turn, creates an “air gap” or “bubble” in

the bypass. Upon start-up the next morning, this air bubble will enter the pump,

causing it to lose prime and all flow will stop—immediately causing the chiller to

shutdown!

If the valve cannot be installed below the tower operating water level, the

bypass must be piped to the tower basin. Thus, while the drain down will still

occur, the resulting air bubble cannot enter the pump. (An alternative would be to

install a check valve in the suction piping between the bypass connection and the

tower basin. But, over time, the check valve will probably leak and the problem

will return.) In this configuration, however, the basin water, at a low temperature,

is supplied to the condenser, so chiller start-up problems may result. Figure 12.2

illustrates the correct bypass valve location in the condenser water piping.

For proper bypass valve installation and operation, the valve should be

located and piped as follows:

1. Locate the valve as close to the condenser as possible.

2. The valve and all piping should be below the tower(s) water operating

level.

3. The valve bypass should be piped directly to the condenser water

supply.

The bypass control valve, ideally, would have linear flow characteristics.

Butterfly valves, which are normally used for bypass duty, have equal percentage

flow characteristics. However, a butterfly valve will have a nearly linear

relationship between “% open” and “% flow” if it is selected to operate with a

maximum 708 rotation and a 33% authority setting.

Additionally, to aid maintenance, the butterfly valves should be the full-lug

type, with operators sized for tight shutoff. Thus, these valves can be used as

isolation valves, also.

Cooling Tower Controls 181

To select the bypass valve on this basis, compute the control valve flow

coefficient Cv from Eq. (12.1).

Cv ¼ PD0:5=Fcdw ð12:1Þwhere PD ¼ required pressure drop across valve at design flow (psig) and

Fcdw ¼ design condenser water flow rate (gpm).

FIGURE 12.2. (a) Cooling tower bypass valve installation: chiller, pump, and

piping located below tower basin level. (b) Cooling tower bypass valve

installation: chiller, pump, and piping located on the same level as the tower basin.

Chapter 12182

PD must be equal to or greater than the sum of the static head and residual

head required by the tower at design flow.

Based on the computed flow coefficient, the correct valve size can be

selected from Table 12.1.

12.2. CAPACITY CONTROL

Tower capacity control is synonymouswith condenserwater temperature control via

manipulation of the airflow through the tower. Condenser water supply temperature

TABLE 12.1. Tower Bypass Butterfly Control Valve Size

Approximate Cv

at 708 rotationPipe size(in.)

Required butterflyvalve size

(in.)

69 3 2136 3 2-1/2201 4 3496 4 4348 6 4611 6 5722 6 6533 8 5784 8 6913 8 8729 10 61,462 10 82,948 10 101,347 12 82,367 12 104,393 12 122,174 14 103,565 14 125,939 14 143,261 16 124,923 16 147,867 16 164,509 18 146,597 18 1610,065 18 186,042 20 168,629 20 1812,535 20 208,789 24 1811,666 24 20


will decrease if the imposed cooling load falls, as indicated by a reduced condenser

water return temperature, and/or the outside air WB temperature lowers. Thus, to

prevent the condenser water from becoming too cold, the tower capacity must be

reduced to maintain the condenser water supply temperature setpoint.

Vapor compression cycle systems have a 1.5–2.5% improvement in

operating efficiency and resulting energy consumption for each 18F decrease in

condenser water supply temperature. Most manufacturers have set the minimum

temperature at 70–758F for centrifugal chillers, though a least one manufacturer

guarantees operation down to 558F condenser water supply temperature. Rotary

screw compressors can operate with a condenser water supply temperature lower

than for centrifugal compressors, but the minimum temperature will vary from

manufacturer to manufacturer. Thus, it is better to establish the condenser water

temperature control setpoint as 70–758F.While a lower condenser water supply temperature will result in lower

chiller energy consumption, the cooling tower fan energy consumption will

increase somewhat for this mode of control. But, studies have shown that chiller

energy savings usually far exceed the additional cooling tower energy consumed

within the temperature ranges discussed here.

Figure 12.3 illustrates a simple condenser water temperature control

sequence, including both tower bypass control and tower fan control.

12.2.1. Fan Cycling

The simplest method of condenser water temperature control is to vary airflow by

cycling the cooling tower fan(s) on and off. It is necessary to establish both a

FIGURE 12.3. Simplified condenser water temperature control system.

Chapter 12184

condenser water supply temperature setpoint and a reasonable throttling range (or

control differential), usually 2–38F. Thus, the fan is started when the supply

temperature rises to setpoint plus half of the differential and stopped when the

supply temperature is reduced to setpoint minus half of the differential. Thus,

with a 758F setpoint and a 38F differential, the fan starts at 76.58F water

temperature and stops at 73.58F.Figure 12.4 illustrates this control process. Note that the “actual”

differential will be greater than the controller differential due to thermal mass in

the system and time delay in initiating and terminating the heat transfer process in

the tower and condenser.

When selecting the controller differential, it is important to pick one that is

reasonable (2–38F). Too great a differential will cause erratic control by allowingtoo much temperature variation, while too small a differential will result in rapid

or “short” cycling of the tower fan(s) and can result in rapid motor failure, drive

damage, or even fan or shaft failure.

A drawback to fan cycling control is that under very light load conditions,

the tower capacity is so great compared to the imposed load that rapid fan cycling

could occur. Thus, as discussed in Sec. 10.7.2, two-speed fan motors or tandem

pony motors are recommended for cooling towers since they allow an

improvement beyond on–off fan cycling control by adding an intermediate

capacity level: half speed. The fan-cycling control concept is the same, except

that a second stage of control is added, as shown in Figure 12.5.

Figure 12.6 illustrates a typical condenser water interlock wiring scheme

for a 2-speed tower fan motor or a main motor and pony motor.

FIGURE 12.4. Condenser water supply temperature control based on tower fan

cycling.


With two-speed or pony motor control, a short time delay (about 15 sec)

must be included when gear drives are used to prevent drive damage that may

result from rapid speed change and the resulting “braking” effect.

12.2.2. Fan Speed Control

In lieu of the “step control” of tower fan(s), cycling, modulating, or

“proportional” control can be achieved through the use of tower inlet dampers,

variable pitch axial fans, or fan speed control.

Dampers, applied to the inlet side(s) of a cooling tower, have proved to be

serious maintenance headaches while, due to their nonlinear control

characteristics at low velocities, providing relatively poor airflow control.

Thus, dampers are generally no longer used in HVAC cooling towers.

The controllable-pitch axial fan concept is widely applied in propeller-

driven aircraft, but has a poor operational history in cooling tower applications.

The mechanical linkages utilized have proven to be unreliable and a constant

maintenance problem. Again, this concept is rarely used in HVAC cooling

towers.

Variable frequency drives (VFDs) have been successfully applied to

cooling tower fan motors so that condenser water supply temperature could be

controlled by changing the fan speed in direct proportion to the change in

FIGURE 12.5. Condenser water supply temperature control based on tower 2-

speed fan cycling.

Chapter 12186

FIGURE 12.6. Dedicated chiller and cooling tower with 2-speed fan interlock

wiring schematic.


condenser water supply temperature. Thus, a 10% increase above setpoint of the

condenser water supply temperature would result in a 10% increase in fan speed,

i.e., proportional control.

With proportional control, there is a linear relationship between the final

control element (the VFD) and the controlled variable (condenser water supply

temperature). The setpoint is typically in the middle of the throttling range, but

the final control element is seldom in the middle of its range, resulting in a control

point that is “offset” from the setpoint. Typically, the control point and setpoint

will be the same at 50% load, but the condenser water temperature will tend to be

lower at light load conditions (less than 50% load) and higher at heavier load

conditions (greater than 50% load). To eliminate offset, an integral function

should be introduced to the proportional control loop. The integral function acts

to “reset” the control point anytime if it drifts off setpoint. This control mode is

called proportional þ integral (PI) and is the recommended mode for fan speed

control.

The application and cost-effectiveness of VFDs has been widely discussed

in the industry and the following guidelines for their application have been

developed:

1. Avoid operating the cooling tower fan at its “critical speed,” or at a

multiple thereof.

2. Be very careful when applying VFDs to gear drives. The braking action

created by the motor as the speed is slowed can cause damage to the

gear drive and the VFD. Also, at low speeds, gear drives may cause

vibration, noise, or gear “chatter.” A separate, electrically driven oil

pump may also be required to provide lubrication at low speeds. (Co-

ordinate these items carefully with the tower manufacturer).

3. Figure 12.7 illustrates the relative energy use by an HVAC cooling with

a 2-speed fan motor vs. the same tower with a VFD. Thus, unless the

tower is very large (40 hp or greater motor size), the savings produced

by the VFD will rarely offset the additional cost for the VFD.

4. Using VFDs on each tower in a multitower system, no matter what the

motor size, is a waste of money since the operating cost savings over

tower staging control will not offset the additional costs for multiple

VFDs. In a case where operation under very light loads is anticipated,

e.g., below 50% of a single tower’s or cell’s capacity, then a VFD on

one tower may be cost effective if (1) the tower is large and (2) this

tower is always the first and last stage in the tower operation sequence.

5. VFDs must have bypass switching (automatic or manual) so that the

tower can be operated even if the drive fails. The fan speed control

system, then, must include provisions to revert to fan cycling control in

this event since the fan can only operate a full speed.

Chapter 12188

Even with VFDs, there is a “minimum” speed required to keep the fan

motor from overheating. If the tower capacity at the minimum fan speed exceeds

the required capacity, the fan control mode must revert to fan cycling, albeit at the

minimum speed.

12.2.3. Tower Staging

In a multiple tower or cell common condenser water system, capacity control is

accomplished by a combination of water flow control and fan cycling.

With increasing load, the automatic isolation valves at each tower or tower

cell should be sequenced open so that flow is provided through all towers before

any tower fan is started. At light cooling loads, maximizing the available heat

transfer surface before running a fan is more energy efficient. Once flow is

established to each tower or cell, the fans can be staged on and off in sequence to

control capacity.

When multiple towers or tower cells utilize 2-speed or pony motors, it is

more energy efficient to operate all cells on low speed before switching any cell

to high speed with increasing load. The incremental power increase associated

with starting an additional fan on low speed is less than 79% of the power

increase from switching a fan from low speed to high speed. (Likewise, on a

FIGURE 12.7. Cooling tower fan power requirement: 2-speed fan cycling

control vs. variable speed control.


falling load, all fans should be switched to low speed before any fan is cycled

off). The use of additional towers or tower cells also increases the heat transfer

surface that is available, decreasing approach and the condenser water supply

temperature.

12.3. MAKE-UP WATER CONTROL

The make-up water to an HVAC cooling tower is controlled by monitoring the

tower basin water level. Two types of devices are used for this duty:

Mechanical float valves are simple float-operated valves that modulate

open to add water as the float falls with the basin water level and

modulate close as the float rises. However, these valves are not reliable

or accurate and are routine maintenance headaches. Thus, they are not

recommended except for the smallest of HVAC cooling towers.

Electronic level sensors and controllers are commonly applied to HVAC

cooling towers. The level sensor has four probes that detect water level

in the basin:

High water level alarm

High operating water level

Low operating water level

Low water level alarm

When the water level drops to the low operating water level, a signal opens

a make-up water solenoid valve to add water to the system. When the

water level rises to the high operating water level, a signal closes the

make-up valve.

12.4. OPERATING SAFETY CONTROLS

The high and low water level alarms from the tower electronic level controller(s)

should be wired to the facility management system or, at a minimum, to both

visual and audible alarm annunciators. (A tower that floods can be annoying, but

a tower that runs dry puts the cooling system out of commission).

For towers with propeller fans, there is always a potential, however small,

that a fan may damage or lose a fan blade. If the fan, now badly out of balance,

continues to operate, vibration damage to the fan, the drive, and the tower

structure can result. To prevent this condition, it is recommended that all

propeller fan towers be equipped with a vibration switch that de-energizes power

to the fan motor in the event of tower vibration. The switch is normally installed

outside the fan cylinder, near the motor, in a weatherproof enclosure (NEMA 3R

or 12 type).

Chapter 12190

13

Condenser Water Treatment

Condenser water treatment is required to prevent or control deposition (scaling),

corrosion, and microbiological fouling of cooling towers, condensers, and piping.

13.1. DEPOSITION CONTROL

Deposition is the term that is used to describe scale and other deposits that may

form on the heat transfer surfaces of a condenser water system. Water has been

called the “universal solvent” and make-up water will always contain dissolved

mineral solids to a greater or lesser extent. The purpose of the deposition control

portion of a condenser water treatment program is to prevent the amount of

dissolved solids from becoming so high that they will begin to be deposited on the

wetted surfaces of the system.

The amount of dissolved solids in make-up water depends on the source of

the water and the geology of the area. However, most regions of the United States

have some sedimentary rock, such as limestone, with which water comes into

contact. The water readily dissolves the calcium and magnesium carbonate, the

major components of limestone, which, in solution, separate into ions of

positively charged calcium and magnesium and negatively charged carbonate

and bicarbonate.

Obviously, the more limestone there is in the local geology, the greater the

amount of dissolved solids, and the “harder” the water. Areas with little limestone

191

will have lower concentrations of dissolved solids and will be “soft” by

comparison. Table 13.1 defines the levels of total dissolved solids relative to

variouswater hardness levels. Since calcium,magnesium, and carbonate are found

all together and they are not easily distinguishable by chemical test, all three are

expressed as CaCO3 (calcium carbonate) for water treatment applications.

Deposits of calcium carbonate on wetted surfaces in a condenser water

system act as insulation and seriously reduce the heat transfer effectiveness.

Deposit thickness can increase to where it reduces flow areas, thus, significantly

reducing flow rates as associated velocity and pressure loss increases. In effect,

significant calcium carbonate deposits will destroy the effectiveness of the

condenser water system and the refrigeration system it supports.

As a portion of the condenser water flow is lost by evaporation through the

cooling tower, the concentration of dissolved solids increases since solids are left

behind as the liquid evaporates. Make-up water, which is added to the condenser

water system to offset the evaporation losses, will add dissolved minerals at a

lower concentration level than the condenser water and, thus, some equilibrium

concentration level will be maintained. However, if this equilibrium

concentration level is high enough that deposition occurs, then a program to

control solids’ concentration at a lower level is required.

The concentration of dissolved solids can be reduced by adding more

make-up water, which has a lower concentration, to the condenser water, which

has a higher concentration. However, to add water to the system, an equal amount

must be removed from the system by blowdown, the intentional “dumping” of

condenser water to drain.

The tendency to form calcium carbonate deposits (scale) is a function of the

temperature, pH, calcium hardness, total alkalinity, and total dissolved solids of

the water:

Temperature: The condenser water supply temperature for HVAC systems

will range from 708F at low ambient wet bulb and light load to 858F at

design wet bulb temperature and load. Most scale-forming dissolved

solids have the unusual property of becoming less soluble as the water

temperature increases. Thus, the greatest potential for scaling occurs at

design conditions.

TABLE 13.1. Water Hardness Scale

Water hardness Concentration of CaCO3 (ppm)

Soft 0–75

Moderately hard 76–150

Hard 151–300

Very hard Over 300

Chapter 13192

pH: Water is made up of hydrogen and oxygen as follows:

Hþ protonated water molecule

OH2 hydroxyl radical or hydroxide

Combining Hþ with OH2 yields H2O, water.

Chemically, pH is a negative logarithmic scale defining the relative

concentration of Hþ, as shown in Table 13.2. Each increase or decrease

of pH by 1 represents a 100-fold increase or decrease in the amount of

acidity or alkalinity (base).

Calcium Hardness: The amount of calcium carbonate in the water, ppm.

Total Alkalinity: The amount of negatively charged ions of hydroxide

(OH2), bicarbonate ðHCO32Þ; and carbonate ðCO2

22Þ in the water, ppm,

and represents the ability of the water to neutralize acid. Alkalinity

occurs in all water with a pH above 4.4.

Total Dissolved Solids: The amount of all dissolved solids (calcium,

magnesium, phosphate, iron, etc.) in the water.

Since 1936, with the introduction of the Langelier Saturation Index (LSI), water

treatment specialists have attempted to “predict” the potential for deposition

based on water chemistry and, thus, establish the required “cycles” to maintain

deposition control.

TABLE 13.2. pH Scale

pH

Hþ concentration

(gmol/L) Relative acid Relative base

14 10214 10,000,000

13 10213 1,000,000

12 10212 100,000

11 10211 10,000

10 10210 1,000

9 1029 100

8 1028 10

7 1027 Neutral

6 1026 10

5 1025 100

4 1024 1,000

3 1023 10,000

2 1022 100,000

1 1021 1,000,000

0 100 10,000,000

Condenser Water Treatment 193

The LSI is computed from Eq. (13.1).

LSI ¼ pHa 2 pHs ð13:1Þwhere pHs is the pH at saturation and pHa is the actual pH.

The pHs is computed on the basis of the water analysis data from Eq. (13.2).

pHs ¼ 12:32 ðlog10Caþ log10TAþ 0:025T 2 0:011TDS0:5Þ ð13:2Þwhere Ca is the calcium hardness, ppm; TA is the total alkalinity, ppm; T is the

temperature, 8C; and TDS is the total dissolved solids, ppm.

LSI generally has a value from þ3.0, which indicates high deposition

potential, to 23.0, which is often used to indicate a high corrosion potential.

However, the LSI prediction for deposition or corrosion is, particularly for

cooling tower systems, often wrong simply because the value of pHa can rarely be

predicted accurately. This value is a function of the properties of the make-up

water, the actual cycles of concentration, the quality of mixing of the condenser

water and make-up water streams, airborne contaminants in the water, etc., and is

almost impossible to accurately predict. In the water treatment industry, the

chemistry of the make-up water is used as the starting point and, as cycles are

increased, the expected value of the actual pH is then computed. While the result

may not be always accurate due to limited number of factors that are addressed, it

does provide a basis upon which to begin.

In 1944, an attempt to improve the prediction of deposition potential was

introduced by the Ryzner Stability Index (RSI) from Eq. (13.3).

RSI ¼ 2ðpHsÞ2 pHa ð13:3ÞTypically, the RSI will have a value of 3.5, indicating high deposition potential,

to 9.0, indicating high corrosion potential. However, the limitation on the

accuracy of predicting pHa is not improved by the RSI and RSI is no more or less

valid as a predictor than the LSI.

Because of the unreliability of both LSI and RSI to accurately predict

deposition or corrosion conditions in cooling tower systems, in the 1970s a new

index called the Puckorius Scaling Index (PSI) was developed. The PSI is based

on correcting the actual pH to match the total alkalinity of the water and is

computed with Eq. (13.4).

PSI ¼ 2ðpHsÞ2 pHeq ð13:4Þwhere pHeq is the adjusted or equilibrium pH computed from Eq. (13.5) as

pHeq ¼ 1:465Log10TAþ 4:54 ð13:5ÞThe range of values of PSI are essentially the same as for RSI.

For corrosion control, condenser water pH should be maintained between 4

and 10, as shown in Figure 13.1. In most cooling tower water treatment programs,

Chapter 13194

the desirable range for pH is between 8 and 9 in order to hold water alkalinity to a

reasonable level (400 ppm or less). However, in metal towers, a pH of 7.0–8.0 is

preferred to help prevent white rust corrosion (see Sec. 13.2.2). Therefore, at least

for most metal cooling towers, the ideal pH range is 7.5–8.5.

Water hardness and alkalinity is a function of the hardness and alkalinity

of make-up water, the amount of evaporation and drift loss from the cooling

tower operation, and the blowdown proposed to yield the desirable pH to

prevent both deposition and corrosion. The term cycles of concentration

defines the ratio of the desired concentration of dissolved solids in the

condenser water to the concentration of dissolved solids in the make-up water,

as follows:

Cycles ðof ConcentrationÞ

¼ Dissolved Solids ðppmÞ in Blowdown=

Dissolved Solids ðppmÞ in Make–up

This relation can be express in terms of water flow by Eq. (13.6).

Cycles ðof ConcentrationÞ ¼ MU=BD ð13:6Þ

FIGURE 13.1. Corrosion rate as a function of water pH. (Courtesy of

BetzDearborn, Inc., Trevose, Pennsylvania.)


where MU is the total make-up water flow, which is the sum of evaporation and

blowdown, gpm, and BD is the blowdown flow, gpm.

Since the amount of evaporation, E, is easily computed as 0.1% of the

condenser water flow rate per degree of range, MU in Eq. (13.6) can be replaced

with the value (E þ BD). Equation (13.6) can then be rearranged to yield Eq.

(13.7):

BD ¼ E=ðCycles2 1Þ ð13:7ÞThus, after the number of cycles is determined based on the make-up and desired

condenser water (blowdown) concentration of dissolved solids, the actual

blowdown requirement can be calculated from Eq. (13.6).

Since drift water loss is not included in these calculations, the

actual required BD flow can be reduced by the amount of drift loss from the

tower.

There are two ways of controlling blowdown in an HVAC cooling tower

system:

1. Constant blowdown with manual adjustment based on periodic water

hardness analysis is the simplest method. However, since the amount of

blowdown is constant, the loss in water and water treatment chemicals

is high and this really represents the most expensive approach.

2. Controlled blowdown is based on continuous monitoring of the water

hardness as indicated by its conductivity. Automatic control minimizes

the waste of water and water treatment chemicals and is the preferred

method, as described in Section 13.4.

Table 13.3 summarizes the water flows associated with a cooling tower for

various cycles of concentration (based on 3.0 gpm/ton flow). From Table 13.3, it

is clear that the amount of make-up water is reduced significantly as the number

TABLE 13.3. Make-Up Water Requirement as Function of Cycles

Cycles

Evaporation

(gpm/ton)

Blowdown

(gpm/ton)

Make-up

(gpm/ton) % Make-up

2 0.0300 0.0300 0.0600 100

3 0.0300 0.0150 0.0450 75

4 0.0300 0.0100 0.0400 67

5 0.0300 0.0075 0.0375 63

6 0.0300 0.0060 0.0360 60

10 0.0300 0.0033 0.0333 55

15 0.0300 0.0023 0.0323 54

20 0.0300 0.0015 0.0315 53

Chapter 13196

of cycles is increased from 2 to 6. However, there is only a further 5% reduction

as the cycles are increased from 6 to 10, and only a further 2% reduction as cycles

are increased to 20. Therefore, in most cooling tower applications, cycles of

concentration are maintained between 5 and 10 (as indicated by the LSI, RSI,

and/or PSI) and deposition inhibitors are added as necessary. While lower cycles

represent loss of more water and treatment chemicals, the amount of treatment

chemicals required tends to go down with cycles, and 5–10 cycles usually

represent a good balance point.

Reduction of hardness by blowdown or even by water softening

pretreatment may not eliminate the potential for deposition in the condenser

water system. In these cases, chemical treatment, i.e., the addition of deposition

inhibitors to the water system, becomes necessary.

When a dissolved salt precipitates and deposits on the wetted metal

surfaces of the condenser water system, it forms a crystal growth that attaches

itself to metal surfaces as it comes out of solution. The most common scale

inhibitors used in condenser water systems are posphonates, which are organic

phosphate compounds, such as HEDP, which function by adsorption on the

crystals as they form, preventing them from attaching to metal. Thus, these

crystals precipitate out of solution, usually in the tower basin.

13.2. CORROSION CONTROL13.2.1. Galvanic Corrosion

Metal corrosion occurs as a result of galvanic action at a negatively charged

“pole” or site on the metal surface. Both anodes, negatively charged sites, and

cathodes, positively charged sites, can be created on the metal due to impurities

in the metal, localized stress, metal grain size or composition differences, or even

scratches on the metal surface. Due to the differences in charges, there is an

electrical potential between the anode and cathode and an electrical current

(electrons) flows from anodes to cathodes, using the surrounding water as a

conductor or electrolyte, as shown in Figure 13.2.

For steel, the anodic reaction is for the iron to give off two free electrons,

becoming positively charged. This positively charged iron then combines two

hydroxyl radicals from the water to form ferrous hydroxide, which combines

further with the water to from ferric hydroxide that, when dehydrated, is iron

oxide or rust. At cathodes, the electrons given off at the anode combine with

water to yield hydroxyl radicals, to balance the reactions.

Corrosion, then is the loss of metal; it literally “dissolves.” Corrosion can

exhibit two characteristics depending on the underlying reason for the anodic and

cathodic sites. General corrosion is wide spread and is caused, usually, by

impurities in the metal or characteristics of the metal or its environment that

results in an overall fouling of the metal surface. Localized corrosion results,


mostly, from scratches, stress, or localized environment and the most common

reason for “metal failure.”

If dissimilar metals with different electrical potentials are used in a

condenser water system, galvanic corrosion is enhanced and the metals simply

corrode faster, particularly at and near the point(s) of contact between the metals.

The first step in corrosion control is to minimize the contact between water

and mild steel materials. All primary wetted surfaces—wet decks for induced

draft towers and basins—should be constructed of stainless steel. This typically

increases tower costs by only about 15% and is generally a worthwhile

investment. Use plastics or fiberglass for the tower casing, wet deck covers,

intake louvers, drift eliminators, and fill. Finally, if mild steel is used for the tower

structural frame, it should be galvanized and, as discussed in Sec. 13.2.2, coated

with an epoxy or polymer final protective coating.

However, the piping in most condenser water systems will be steel and

must be protected from corrosion. This is accomplished by using one or more

treatment programs, as follows.

Passivating (Anodic) Inhibitors: These chemicals form a protective oxide

film on the metal surface which is not only tough, but, when damaged,

quickly repairs itself. Typical chemicals that act as passivating inhibitors

include molybdate, polyphosphates, and orthophosphate. These

chemicals are oxidizers that promote passivation by increasing the

electrical potential of iron. The drawback to the use of molybdate is its

expense, and it is used only when blowdown levels are kept as low as

FIGURE 13.2. Typical steel corrosion chemistry. (Courtesy of BetzDearborn,

Inc., Trevose, Pennsylvania.)

Chapter 13198

possible. Othophosphates should not be used in condenser water systems

containing stainless steel since it will make the metal brittle over time.

Precipitating Inhibitors: At cathodic sites, the localized pH at the site is

increased due to the higher concentration of hydroxide ions that are

being produced. Precipitating inhibitors form complexes that are

insoluble at the higher pH and, thus, precipitate out of the water. Zinc is a

good precipitating inhibitor. Molybdate will also act as a precipitating

inhibitor and, thus, can serve as a corrosion inhibitor using two

mechanisms.

Adsorption Inhibitors: These are organic compounds containing nitrogen,

such as amines, or sulfur or hydroxyl groups. Due to the shape, size,

orientation, and electrical charge of the molecule, they will attach to the

surface of the metal, preventing corrosion. Their drawback is that they

form thick, oily surface films that reduce heat transfer capability.

Each program is designed for particular condenser water pH range and water

chemistry, as summarized in Table 13.4.

To protect copper in heat exchanger tubes and piping from corrosion,

aromatic triazoles, such as benzotrizole (BZT) and tolyltriazone (TTA), are used

in most condenser water treatment systems. These compounds bond with the

cuprous oxide on the metal surface and protect it.

13.2.2. White Rust

Since the late 1970s, there has been a significant increase in the use of galvanized

steel cooling towers, particularly in the HVAC market. With the advent of new

cooling tower water treatment polymers and corrosion inhibitors, most condenser

water treatment programs now utilize little or no acid addition, and operating pH

levels have increased from 6.5–7.5 to as high as 9.5 (with the new alkaline

polymer approach). As the technology of the water treatment industry has

changed, the corrosion of galvanized steel has now become a major concern.

The term white rust refers to the premature, rapid loss of galvanized coating

on cooling tower metal surfaces. White rust is evidenced by a white, waxy,

nonprotective zinc corrosion deposit on wetted galvanized surfaces. This rapid

TABLE 13.4. Water pH and Chemistry for Corrosion Programs

Program Water pH and chemistry

Zinc 7.5–8.5

Molybdate or molybdate/zinc 7.5–9.5

Orthophosphate 7.5–8.5, with a phosphate deposition inhibitor

Organic adsorption 7.5–9.5, with 300–500 ppm alkalinity


loss of galvanizing results in the corrosion of the underlying steel and, instead of

tower systems that will last 20–25 years, equipment will have drastically

shortened life spans.

Many believe that the change to the dry kettle method of galvanizing, with

continuous sheet steel (see Chap. 10), has had an effect on the increased

formation of white rust. Initial research has shown that the levels of aluminum

and lead in the galvanizing have changed. With the wet kettle method, lead levels

were 0.60–1.0% and aluminum levels were 0.005%. Levels of lead have dropped

to 0.05% and levels of aluminum have increased to 0.40% since going to the

continuous sheet galvanizing. This increase in aluminum and decrease in lead is

believed to help increase the brightness of the metal surfaces, making a better-

looking product. The higher aluminum percentage is also necessary for better

bonding of the zinc coating to the steel.

The cooling tower manufacturers do not believe that changes in the

galvanizing process or in the lead and aluminum levels deposited are responsible

for the increase of white rust. The industry has stated that aluminum levels have

not changed and that the galvanized coating is actually more than twice as thick

(2.35 oz of zinc per ft2 of steel sheet) as they were 20 years ago.

However, it can be documented that cooling water programs operated with

a pH range of 8.0–8.5 for many years without having white rust problems. It can

also be documented that new towers added to existing systems, using the same

make-up water, chemical treatment, and controls have developed white rust,

while the original units have not. Therefore, white rust does not appear to be a

water quality and/or water treatment problem.

So while the argument continues, and based on current information, it is

well established that white rust may form if the following conditions exist:

1. The galvanized coating is not properly “passivated” when the tower is

placed in service. (Passivation is a process that allows the zinc coating

to develop a natural nonporous surface of basic zinc carbonate. This

chemical barrier prevents rapid corrosion of the zinc coating from the

environment, as well as from normal cooling tower operation. The

basic zinc carbonate barrier will form on galvanized surfaces within 8

weeks of tower operation with water of neutral pH (6.5–8.0), calcium

hardness of 100–300 ppm, and alkalinity of 100–300 ppm).

2. Condenser water is maintained at pH above 8.0.

3. High condenser water alkalinity (above 300 ppm).

4. Low condenser water calcium hardness level (below 100 ppm).

5. The lack of phosphate-based corrosion inhibitor in the condenser water

treatment program.

For most galvanized metal HVAC cooling towers, white rust will occur if not

prevented by the following steps:

Chapter 13200

1. Provide a secondary barrier coating on all wetted surfaces, such as

epoxy or polymer finish for a new tower or coaltar (bitumen) on an

existing tower. An even better approach is to specify new towers to

have wetted surfaces, such as basins and wet decks, to be constructed of

stainless steel. This option is normally available for only a 10–20%

cost increase.

2. Run the cooling water treatment program at a pH between 7.0 and 8.0,

which may require pH control.

3. Make sure the galvanized tower is properly passivated upon system

start-up. Where white rust has occurred, the metal can be “re-

passivated” by treating the surface with a 5% sodium dichromate in

0.1% sulfuric acid, brushing with a stiff wire brush for at least 30 sec,

then rinsing thoroughly.

4. Incorporate a phosphate-based product into the water treatment

program, along proper dispersants.

13.3. BIOLOGICAL FOULING CONTROL13.3.1. Biological Fouling

Biological fouling results from bacteria, fungi, zooplankton, and phytoplankton

or algae introduced through make-up water or filtered from the air passing

through an HVAC cooling tower. “Fouling” results when these micoorganisms

grow in open systems rich in oxygen (an aerobic process) and form slime on

the surfaces of the tower, piping, and heat transfer surfaces of the condenser

water system. Slime is an aggregate of both biological and nonbiological

materials. The biological component, called the biofilm, consists of microbial

cells and their byproducts. The nonbiological components consist of organic

and/or inorganic debris in the water that has become adsorbed or imbedded in

the biofilm layer.

The impact of biological fouling is twofold: the slime acts as an insulator

and reduces heat transfer efficiency in the system, and microbial activity within

the slime can accelerate corrosion by creating a localized oxygen-rich

environment that accelerates oxidation.

A very special case of biological fouling in cooling towers is the bacteria

Legionella, which is discussed in Sec. 13.3.2.

The number one method of controlling biological fouling is to keep cooling

towers clean. At least twice during the cooling season, the tower should be

drained, scrub cleaned, and allowed to fully dry before refilling. Then, the use of

chemical treatment will complete the control chore.

There are two kinds of antimicrobial chemicals or biocides used in cooling

tower water treatment programs to control biological fouling: oxidizing and

nonoxidizing.


Oxidizing chemicals include chlorine, bromine, and ozone that oxidize or

accept electrons from other chemical compounds. Used as antimicro-

bials, these chemicals react directly with the microbes and degrade

cellular structure and/or deactivate internal enzyme systems. They

penetrate the cell wall and disrupt the cell metabolic system to “kill” it.

(Warning! Oxidizing chemicals, particularly chlorine, can react with

steel, including stainless steel, and cause rapid corrosion. To prevent

this, concentrations of these chemicals must be kept low, ideally to less

than 0.7 ppm. Oxidizing chemicals must be introduced into the

condenser water system in such a way to be rapidly dispersed to

prevent localized high concentrations.)

Nonoxidizing antimicrobials attack cells and damage the cell membrane or

the biochemical production or use of energy by the cell, resulting in its

death, and are sometimes referred to as “surface-active” biocides.

Typical nonoxidizing biocides include isothiazolines, gluteraldehyde,

MBT, and polyquat.

Microbials in condenser water systems can become resistant to a single

method of attack, or some microbials may be more or less immune to one type of

attack. Therefore, it is recommended that both types of treatment chemicals be

used (oxidizing and nonoxidizing), either blended together or in alternating

treatment patterns, as indicated by periodic water testing results. The key to a

successful biological treatment program is maintaining adequate chemical

treatment levels at all times via continuous feed of antimicrobials into the

condenser water system.

13.3.2. Legionella Control

In 1976, 34 attendees at an America Legion Convention in Philadelphia died

from a pneumonia-like disease that was later traced to the then-unknown bacteria

that we now call Legionella. Since this initial outbreak (which was really

preceded by earlier events in Austin, MN in 1957 and Washington, DC in 1964),

the recognition of Legionella as a serious problem has grown significantly.

According to the U.S. Centers for Disease Control (CDC), Legionnaires’ Disease

infects approximately 25,000 people in the United States each year and 10–15%

of these cases are typically fatal. This equates to 3000–7000 deaths attributable

each year to Legionella.

Legionella is a bacteria that is common in surface waters, including lakes,

rivers, etc. The bacteria survive routine water treatment and low concentrations are

introduced into most potable water supplies. Legionella thrives in water

temperatures between 68 and 1228F, with optimal growth occurring between 95

and 1158F. Low pH and high levels of aquatic growth (microbiota, amoebae, algal

Chapter 13202

slime, etc.) enhance bacteria growth.Water temperatures above 135–1408F kill thebacteria.

At these temperatures, the ideal habitats for Legionella include cooling

tower and evaporative condenser systems, where temperatures typically range

from 70 to 1008F. Other potential breeding grounds for Legionella include

domestic hot water systems in schools and hospitals, humidifiers, spas or

whirlpools, and even vegetable misters in supermarkets.

The major mechanism for infection by Legionella is via inhalation of

aerosolized water droplets or particles containing the bacteria. Cooling tower or

evaporative condenser sprays introduce aerosolized water droplets and, therefore,

represent prime mechanisms for infecting humans. There is no evidence that

drinking water with the bacteria in it will cause disease, nor can the disease be

passed by human-to-human contact.

There are two types of disease caused by Legionella: “Legionnaires’

Disease” is a severe form of pneumonia, while “Pontiac Fever” is a nonfatal flu-

like illness. Legionnaires’ Disease symptoms can vary from a cough and low

fever to rapidly progressive pneumonia, coma, and death. Symptoms occur

typically within 3–9 days after exposure.

To monitor and control Legionella, the following steps are recommended:

1. Test for Legionella in cooling tower water. While there is an academic

debate over the cost-vs.-benefit of routinely testing for Legionella, it is

dumb (from both an ethical and a liability point of view) to ignore any

potentially life-threatening condition in a facility. Testing must be

specific for Legionella: “total bacteria” tests promoted by some water

treatment companies are inadequate since there is no correlation

between total bacteria and Legionella concentrations.

2. If Legionella is found, reduce or eliminate the bacteria concentration.

Cooling towers (and evaporative condensers) can be decontaminated

by slug chlorination with 50 ppm of free residual chlorine, along with a

dispersant. Then maintain 10 ppm of free residual chlorine for a least

24 hr, while maintaining the water pH at 7.5–8.0. Finally, drain the

system and repeat the process.

3. Cooling tower systems should be drained and flushed twice each year.

All surfaces should be cleaned and allowed to air dry before reuse.

Continuous-feed water treatment systems for biocides are required to

maintain consistent concentration levels.

4. Keep up with the latest information on Legionella. ASHRAE has

published Guideline 12-2000, Minimizing the Risk of Legionellosis

Associated with Building Water Systems, and CTI has also published a

guideline specifically addressing cooling towers.


13.3.3. Microbiologically Induced Corrosion (MIC)

MIC is caused by sulfate-reducing bacteria (SRB) in the water and is usually

evidenced by reddish or yellowish nodules on metal surfaces. When these

nodules are broken, black corrosion by-products are exposed and a bright silver

pit is left in the metal. A “rotten egg” smell when the nodule is broken is also

evidence of SRB corrosion.

SRBs obtain their energy from the anaerobic reduction of sulfates that are

available in most water. The bacteria contains an enzyme that enables it to use

hydrogen generated at a cathodic site to reduce sulfate to hydrogen sulfate and act

like a cathodic “depolarizing agent.” Iron corrosion by this process is very rapid

and, unlike rusting, is not self-limiting.

Once SRBs begin to grow in a system, they are very difficult to eliminate.

Thus, a preventative program is far more effective than a clean-up program:

1. Keep the system clean through sidestream filtration and regular

cleaning.

2. Prevent contamination by oils and/or grease. Even very small amounts

can cause problems.

3. Eliminate potential bacteria sources (such as bathroom and kitchen

vents, diesel exhaust, etc.) near cooling towers.

4. Run the condenser water pump as much as possible. Do not allow

stagnant conditions to exist.

The condenser water system should be regularly tested for SRBs in accordance

with ASTM Standard D4412-84 (1997) Standard Test Methods for Sulfate-

Reducing Bacteria in Water and Water-Formed Deposits. Normally, control of

SRBs can be accomplished with the housekeeping measures outlined above,

coupled with the use of an oxidizing antimicrobial. However, if SRBs do become

established in the system, biocides are no longer effective and a special clean-out

program must be designed for each system.

13.4. WATER TREATMENT CONTROL SYSTEMS

An “automatic” condenser water treatment control system must control four

elements:

1. Amount of blowdown

2. Addition of deposition (scale) inhibitor

3. Addition of corrosion inhibitor

4. Addition of antimicrobials (usually two of which are required)

Figure 13.3 illustrates a typical cooling tower water treatment system and its

components.

Chapter 13204

FIG

URE13.3.

Typical

automatic

condenserwater

treatm

entcontrolsystem

schem

atic.


The amount (cycles) of blowdown is controlled by monitoring and

maintaining a specific conductivity setpoint for the condenser water.

Conductivity in water is a function of the amount of dissolved solids in the

water, since the dissociated salts serve as an electrolyte. Therefore, conductivity

can be measured and a desirable setpoint established to maintain the desired

amount of dissolved solids.

Blowdown is initiated by opening a blowdown solenoid valve on the

condenser water return line (water returning to the tower). Blowdown can be

controlled by “continuous bleed” that modulates the control valve and amount of

blowdown in direct proportion to the level of conductivity or by “intermittent

bleed” by opening and closing a two-position valve to maintain conductivity

within established high and low limits.

The injection of deposition and corrosion inhibitors and dispersants are fed

in proportion to the make-up water or blowdown water flow rate, measured by

flow meters in both flow streams, by modulating metering pump speeds.

Antimicrobials are fed, typically, on the basis of a constant rate of injection

established from periodic water testing. The two types of antimicrobials

(oxidizing and nonoxidizing) can be alternately fed on a fixed time schedule, feed

continuously together, or one type can be fed continuously and the other added

manually to “shock” the system.

13.5. ALTERNATIVE WATER TREATMENTMETHODS

13.5.1. Sidestream Filtration

HVAC cooling towers are excellent “air washers.” Pollen, dust, microbes,

leaves, and other debris in the air are readily trapped and removed by the

water and deposited in the basin to form “sludge,” which can foul heat

transfer surfaces. In the majority of cases, regular basin cleaning will control

the problem, but in many urban and/or industrial areas, the amount of sludge

formed during normal tower operation may be so great that specific control

measures must be installed.

The most common general fouling control method is the use of side-steam

filtration. Here, a portion of the condenser water flow is diverted through a filter

for removal of dirt and suspended solids. Typically, filters are sized so that the

entire water volume is filtered each hour. Thus, it is necessary to determine the

total system water volume contained in the tower basin and wet deck, condenser

water piping, condenser(s), etc., and divide that volume, in gallons, by 60min to

establish the required filter flow rate, in gpm.

The most common (and economical) type of sidestream filter is the sand

filter with backwash, much as that used for swimming pool applications. These

filters will remove suspended solids with 50mm and larger particle size.

Chapter 13206

13.5.2. Ozone Treatment

Ozone is a condenser water antimicrobial that eliminates the need for chemical

treatment for microbiological fouling.

Ozone (O3) is an unstable form of oxygen (O2) that has a relatively short

half-life, usually less than 10min. Ozone is a powerful biocide and virus

deactivant and will oxidize many organic and inorganic compounds.

When reacting with another molecule, the ozone molecule is destroyed,

producing carbon dioxide, water, and a partially oxidized form of the original

reactant molecule. Any residual ozone will decompose and recombine as oxygen,

producing no toxic or carcinogenic byproducts.

Ozone can be produced by UV radiation or by high voltage corona

discharge. However, the UV radiation method is very inefficient and the high

voltage corona discharge method is normally applied. This is accomplished by

passing a high voltage AC current (6–20 kV) across a dielectric gap of 10–

15mm, through which dry, compressed air is injected. As the compressed air

passes through the voltage field, the oxygen molecules dissociate and form single

oxygen atoms, some of which combine to form ozone (about 1–4%). The

compressed air and ozone mixture is then injected into the condenser water,

typically at the tower basin.

Filtration downstream of the ozone injection point is required since the

ozone generation process will produce mineral deposits that must be removed

to prevent fouling. As the cycles of concentration increase, calcium and

carbonate ions in solution will precipitate out as calcium carbonate and settle

in low water velocity areas of the system, typically in the tower basin.

Ozone treatment cannot be used with water with excessive hardness

(500 ppm or higher calcium carbonate) or with sulfates greater than 100 ppm.

Also, because of its short life, ozone should not be used in large systems or

systems that have long piping runs that would require long residence times to get

complete coverage. Water temperatures above 1108F eliminate the effectiveness

of ozone treatment.

Ozone exposure for workers is also a consideration. USOSHA has

established an exposure limit of 0.1 ppm in the air over an 8-hr exposure period.

Monitoring of ozone concentration is required if towers are located at ground

level, adjacent to personnel.

The potential benefits from ozone treatment are twofold:

1. Effective antimicrobial treatment without the use of expensive

chemicals

2. Reduced blowdown since the ozone generation process precipitates out

dissolved solids


13.5.3. UV Treatment

Ultraviolet light can be used in lieu of chemicals as antimicrobial treatment for

condenser water to prevent microbiological fouling. Typically, a portion of the

condenser water flow (3.5–10%) is diverted through the UV treatment equipment.

UV systems are offered in several configurations. The UV lamps that are

used have two different wavelengths, 185 and 254 nm. For cooling tower

applications, the 254 nm lamp has been found to be the most effective. This

wavelength is effective for germicidal applications and does not break down the

other chemicals used for deposition and corrosion control.

Aside from wavelength, lamp intensity is important. Lamps are typically

available at both 39 and 65W. Based on the system size and the flow rate, the

smallest number of lamps will result in the lowest maintenance (replacement) costs.

Each UV lamp is installed with a quartz sleeve, which must include a

“wiper system” to keep the sleeve clean so that the maximum output of UV light

is provided.

The UV system is typically piped in a “side stream” arrangement on the

condenser water pump discharge. A filter must be installed ahead of the UV

system to minimize the fouling of the lamp quartz tubes. Filtration can be

cartridge, bag, sand, or cleanable mesh type, but they must be rated to stop

particles 50mm and larger in size.

13.5.4. Magnetic Treatment

Every few years, salesmen make the rounds to tower owners and designers and

talk about the “wonders” of magnetic treatment for deposition control in

condenser water systems. Their product consists of strap-on magnets installed

around condenser water piping.

Promoters of these devices claim that dissolved solids, which dissociate

into charged ionic salts in solution, can be easily removed by allowing them to

pass through a magnetic field. Tests by independent sources have proved

conclusively that magnetic treatment has no effect on deposition rates in

condenser water systems. Magnetic treatment is simply a scam and, when

encountered, should be immediately rejected as a viable water treatment method.

13.6. TREATMENT FOR WOODEN TOWERS

For very large cooling towers, e.g., 1000–1500 tons and larger, wood is still

widely used for cooling towers. Wood is composed of three major elements:

Cellulose, which forms long fibers in wood and gives it its strength

Lignin, the soft material that acts as a cementing agent for the cellulose

Extractives, the natural compounds that enable wood to resist decay.

Chapter 13208

Wood deterioration can result from chemical attack, biological attack, and/or

physical decay due to (elevated) temperature degradation or rupture of wood cells

by crystallization of dissolved solids in the condenser water. Physical decay is

relatively uncommon in HVAC applications.

Chemical attack commonly results in delignification of the wood, resulting

in wood that has a white or bleached appearance and a fibrillated surface.

Delignification is usually caused by oxidizing agents and alkaline materials and is

particularly severe when high chlorine residuals (more than 1 ppm free chlorine)

and high alkalinity (pH of 8 or higher) occur simultaneously.

The biological organisms that attack cooling tower wood are those that use

cellulose as a source of carbon for growth. These organisms degrade the cellulose by

secreting enzymes that convert the cellulose into compounds they can then absorb as

food. This type of attack deletes the cellulose and leaves a residue of lignin. The

wood appears dark, loses much of its strength, and becomes soft and spongy.

Biological attack is of two basic types: surface rot and internal decay.

Surface rot is easily detected and can be repaired, but internal decay, usually

found in the plenum area, cell partitions, decks, fan housing, supports, and other

areas that are not continuously flooded, is more difficult to detect simply because

the exterior of the wood will display little or no sign of the rot.

Water treatment and preventative maintenance is the only way to prevent

wood deterioration:

1. For flooded areas of the tower, use nonoxidizing antimicrobials to

control slime and prevent biological attack.

2. Internal decay is the most common problem for nonflooded areas of a

tower and these areas should be thoroughly inspected at least twice

each year. Test for soundness with a blunt probe (long, thin

screwdriver) and, when suspected, a sample of wood should be

examined microscopically to detect internal microorganisms. Any

infected wood should be replaced immediately with pretreated wood to

prevent the spread of the infection to healthy wood. (Periodic spraying

with an antifungal treatment may be helpful in reducing biological

attack, but it must be done very thoroughly and on a regular schedule,

otherwise it is a waste of time and chemicals.)

13.7. CHEMICAL STORAGE AND SAFETY13.7.1. Spill Control

Chemicals for water treatment are usually purchased in bulk and, thus, there is a

potential for a chemical spill when the drums are moved or opened. To eliminate

this potential, either of two measures may be used:


1. Purchase chemicals in double-wall containers that have integral spill

containment. While more expensive than simple drums, these

containers provide protection from internal rupture and spillage while

filling or emptying. The drawback to them is that there is no protection

from a puncture that penetrates both the outer and inner walls of the

container.

2. Construct a “spill containment” reservoir for use with conventional

single wall drums. This reservoir usually consists of a concrete curb

enclosing the chemical storage area. The volume of containment must

be greater than the single largest container and a “sump” or depressed

area must be provided, with the floor sloped to it, for collection of any

spilled liquid for pump-out. Grated flooring can be installed on top of

the curb and the drums stored on it, or ramps can be installed on both

sides of the curb and the drums sit on the floor inside the containment

area.

13.7.2. Safety Showers and Eye Wash Stations

USOSHA Standard 29 CFR 191.151(c) requires eyewash and shower equipment

for emergency use where the eyes or body of an employee may be exposed to

injurious materials. Typically, condenser water treatment chemicals would be

considered to be potentially injurious. This safety equipment must be designed,

installed, tested, and maintained in accordance with ANSI Z358.1. This standard

requires that eyewash and shower equipment be located based on the estimated

time of travel of a person with compromised vision. Equipment must be in

accessible locations that require no more than 10 sec to reach. Combination units

(shower plus eyewash) must be designed and connected to the water system so

that both units can be operated simultaneously by the same user. Flow is required

within 1 sec of activation and stay-open valves are required so that the user is free

to use his/her hands to remove clothing or hold the eyelids open.

ANSI Z358.1 mandates the delivery of tepid water—moderately warm or

lukewarm—for 15min at 20 gpm in emergency showers and 0.4 gpm in

emergency eyewash units. The standard does not specifically define tepid, but the

tepid range is generally considered to be 78–928F, based somewhat on the

normal surface temperature of the human eye. There is no grandfathering

provision in the standard; an existing cold water shower or eyewash is simply no

longer compliant and must be changed.

Tempering water for emergency showers and eyewash applications is a

challenging problem. When selecting an automatic thermostatic mixing valve for

a tepid emergency shower or eyewash system, consider the following:

1. Combination shower and eyewash units are frequently used. The

shower has a high flow rate (about 20 gpm), while the eyewash has a

Chapter 13210

low flow rate (about 0.6 gpm). The mixing valves must be capable of

both high and low flow control.

2. A “temperature spike” can occur when large, equal quantities of hot

and cold water are mixed instantaneously, creating a sharp rise in the

outlet temperature. In domestic water applications, the piping system

has time to flatten a temperature spike. In an emergency, however,

valves are typically installed very close to the fixture and this time

delay is lost. The mixing valve must respond very quickly to flatten any

temperature spike as it begins.

3. The majority of domestic water mixing valves are designed to restrict

flow in the event of hot or cold water failure. However, at an emergency

shower, flow reduction is unacceptable. In an emergency, cold water

flow is better than no water flow (while in a domestic water situation

that would be unacceptable).


14

Special Tower Considerations

14.1. BASIN AND OUTDOOR PIPING FREEZEPROTECTION

Except in northern climates, where towers are often drained and secured for the

winter, towers will stay full of water and ready for use year-round. However,

even in mild climates, there will be periods when the ambient temperature drops

below 328F and basin water freezing is probable.

To prevent that problem, basin heaters are normally installed. These heaters

are most commonly electric resistance elements, but steam coils or steam ejectors

can also be used.

The heat loss from a cooling tower basin, with 408F water, can be estimated

from the data provided by Table 14.1. Thus, a 120 wide by 200 long steel cooling

tower basin with a 1200 of water depth would have estimated heat losses for 208Foutdoor temperature computed as follows:

Surface loss: 120 £ 200 £ 80Btu=hr sf ¼ 19,200Btu=hr

Sides loss: 10 £ ½ð2 £ 120Þ þ ð2 £ 200Þ� £ 40Btu=hr sf ¼ 2560Btu=hr

Bottom losses: 120 £ 200 £ 40Btu=hr sf ¼ 9600Btu=hr

Total: ¼ 31,360Btu=hr

212

For an electric immersion heater, each kW produces 3413Btu/hr. In this case, a

9 kW heater would be required to maintain the tower basin at approximately 408Fwith a 208F outdoor air temperature.

Electric heaters are typically used to maintain the basin water temperature

above freezing, since they are relatively inexpensive. But, electric heat suffers

from two problems: First, it is costly to operate and, second, since the heating is

localized, poor mixing and over- or under-heating in different areas of the basin

can result.

Steam injection heaters can be used if steam is available. These heaters are

submerged in the basin and, when heating is required, a two-position automatic

valve is opened and steam is injected through a venturi or nozzle that sets up a

convective current in the water ensuring relatively good mixing. Steam coils,

submerged in the basin water can also be used.

Outdoor condenser and make-up water piping must also be protected from

freezing. Typically, this is accomplished by wrapping the pipe with electric heat

tracing elements (though, again, steam tracing can be used) before the piping is

insulated. For cooling tower piping applications, the criteria from Table 14.2 can

be used to prevent pipe freezing. (In most of the southern United States, where

outdoor winter design temperatures are 208F or higher and there are no prolonged

periods at these low temperatures, heat tracing can be safely eliminated for piping

larger than 400.)

14.2. WATERSIDE ECONOMIZER CYCLE

Commonly, HVAC systems are designed to utilize an airside economizer cycle to

provide “free” cooling when the outdoor temperature is at or below the required

supply air temperature. However, there are a number of secondary system types

in which the airside economizer cannot be applied, such as with fan coil units. In

these cases, the waterside economizer cycle may have application.

TABLE 14.1. Approximate Cooling Tower Basin Heat Loss

Heat loss basin sides and bottom (Btu/hr sf)Outdoor air

temperature

(8F)

Approximate

water surface

heat loss (Btu/hr sf) Steel or fiberglass Concrete (800 thick)

30 50 20 15

20 80 40 25

10 110 60 35

0 130 80 50

210 160 100 60

220 185 120 70

230 210 140 80

Special Tower Considerations 213

This cycle utilizes a cooling tower to produce cold water when the ambient

wet bulb temperature is 508F or lower. At 508F WB, a typical cooling tower will

produce 40–50% of its design capacity at an approach of about 58F. Thus, a500-ton cooling tower will deliver approximately 250 tons of cooling with a 458Fsupply water temperature at 508F ambient wet bulb. But, in most comfort cooling

applications, the need for dehumidification is reduced during the heating season

and a 5–108F increase in the chilled water supply temperature can be tolerated.

Thus, the number of hours of available free cooling can be significantly

increased.

Obviously, this cold water can be used to provide cooling without operating

a water chiller. However, to separate the “dirty” tower water from the “clean”

chilled water, a heat exchanger is used, at only a loss of 2–38F in supply water

temperature.

Two types of heat exchangers may be used for waterside economizer

applications:

1. Plate and frame heat exchangers, often called simply “frame heat

exchanges” and shown in Figure 14.1, consist of thin corrugated plates

separated by rubber or neoprene gaskets and mounted in a rigid frame.

The two water flows are arranged counter to each other on opposite

sides of each plate. Close temperature approach, 1–28F, is obtained by

having a very large heat transfer area in a relatively small package.

2. High efficiency shell and tube heat exchangers provide high heat

transfer area by (a) using a large number of small diameter enhanced

surface tubes and (b) increasing the tube length with a U-shaped

TABLE 14.2. Electric Heat Tracing for

Outdoor Piping

Pipe size (in.) Watts/linear foot of piping

1/2 1.6

3/4 1.7

1 1.9

1-1/4 2.0

1-1/2 2.3

2 2.9

2-1/2 3.4

3 4.5

4 4.9

6 7.1

8 and larger 0.68/in. pipe diameter

Chapter 14214

configuration. Figure 14.2 shows a typical high efficiency shell and

tube heat exchanger. An approach of 28F can be obtained with this type

of heat exchanger.

Shell and tube heat exchangers have lower cost, less weight, and more versatile

configurations than plate and frame heat exchangers. They are also easier to

insulate. But, closer approach (higher efficiency) can be obtained with the plate

and frame heat exchanger.

While the economizer can be applied to a single tower system, most energy

codes require that it be an “indirect” system, shown in Figure 14.3, which allows

the chiller and the economizer to operate concurrently. Since chillers normally

will not operate at a 458F condenser water supply temperature, bypass

temperature control becomes critical and often fails. However, where multiple

chillers and cooling towers are available, one tower can be isolated for

economizer duty, leaving at least one other chiller and tower available to

supplement the economizer cooling capacity, as shown in Figure 14.4.

The energy consumption by the waterside economizer for the pumps and

cooling tower is only about 15–20% of the energy consumption required by

mechanical cooling.

FIGURE 14.1. Typical plate and frame heat exchanger installation. (Courtesy of

Alfa Laval Thermal, Inc., Richmond, Virginia.)


14.3. NOISE AND VIBRATION

Noise is simply objectionable sound and “community noise” and “noise

pollution” are terms applied to sound generated on one property, but deemed

objectionable by listeners on adjacent properties. Along with trucks and cars,

trains, the neighbor’s stereo, and leaf blowers, cooling towers are often the source

of community noise problems.

In 1972, the U.S. Congress passed the Noise Control Act and authorized the

EPA to begin a process of establishing emission standards for outdoor noise

sources. However, funding for this effort was deleted in 1982 and, for all practical

purposes, community noise control efforts have been delegated to the state and

local governments.

Numerous states and municipalities have property line sound limits that

must be met. Under this type of regulation, maximum sound pressure levels

(dBA, see Sec. 6.1), measured at adjacent property lines, are established. Typical

maximum allowable sound pressure values are 50–60 dBA in residential areas

and 60–70 dBA in industrial areas.

More stringent standards have been developed by the World Health

Organization (WHO). The WHO standards are being promoted by numerous

antinoise groups and their adoption by at least some states and municipalities is

FIGURE 14.2. Typical shell and tube heat exchanger. (Courtesy of the


Chapter 14216

considered but a matter of time. WHOmakes the case that the effect on the health

and safety of the listener by a combination of noise “events” is related to the

combined sound energy of those events. Thus, they introduce the term “LAeq,T”

to define A-scale sound pressure level (dBA) averaged over a time period T. As

shown in Table 14.3, these standards establish maximum LAeq,T limits at or

within various types of occupancies or environments.

Typical cooling tower noise emission is in the range of 65–85 dBA

measured at (typically) 5 ft from the tower, and each tower manufacturer can

provide specific sound emission data for the required operating conditions.

For most propeller fan towers, the manufacturers have developed “low

noise” options, such as adding extra fan blades so the fan can be operated at lower

speed, etc. Centrifugal fans are quieter than propeller fans and may be a good

option if community noise is expected to be a problem.

There are two caveats relative to manufacturer-provided sound data:

1. The noise data provided is typically based on a single tower or tower

cell. Where multiple towers or tower cells are utilized, the manufacturer

must be queried for cumulative effect sound data. In the field, fan

speeds on multiple towers or cells must be as nearly identical as

possible to prevent annoying “out of phase” sound differences.

FIGURE 14.3. Piping schematic for waterside economizer with a single chiller

and cooling tower.


2. The noise emission from the tower may not be uniform in all directions.

For crossflow towers, the inlet side(s) will be noisier than the enclosed

side(s). The manufacturer data must define the tower orientation for

each noise emission value.

The basic sources of cooling tower noise include the following:

1. Fans

2. Air movement and turbulence through the tower

FIGURE 14.4. Piping schematic for waterside economizer with multiple chillers

and cooling towers.

Chapter 14218

3. Motor and/or drive

4. Structural or casing vibration

Fan noise is related to the fan design: Centrifugal fans are inherently quieter than

axial flow fans. Axial fan noise can be reduced by (1) reducing the fan speed

and/or (2) increasing the number of fan blades.

The tower location and orientation can have a pronounced effect on the

sound condition. Always orient a quiet side of the tower toward any sound

reflecting wall or side of a building or toward adjacent property. Locate towers at

least 20 ft away from any sound reflecting surface.

Sound dissipates over distance. As shown in Table 14.4, the amount of

sound attenuation varies with the multiples of distance from the point of sound

measurement (distance from the rated noise measuring point). At a distance 10

times greater than the sound rating distance from the tower, approximately 20 dB

attenuation is expected. Thus, a tower with a rated noise level of 86 dB at 5 ft

TABLE 14.3. World Heath Organization Community Noise Limit Guidelines

Environment Critical health effect(s)

LAeq

(dBA)

T

(hr)

Outdoor living area Serious annoyance 55 16

Moderate annoyance 50 16

Dwelling (living

areas, inside)

Speech intelligibility

and moderate annoyance

35 16

Bedrooms (inside) Sleep disturbance 30 8

Bedrooms (outside,

window open)

Sleep disturbance 45 8

Schools (classrooms,

inside)

Speech intelligibility,

disturbance of information

extraction, message

communication

35 During

class

Schools (playground,

outdoors)

Annoyance 55 During play

Hospital (patient

rooms, inside)

Sleep disturbance 30 8 (night)

16 (day/

evenings)

Industrial, commercial,

shopping, and traffic

areas (inside and

outside)

Hearing impairment 70 24

Outdoors in parkland

and conservation

areas

Disruption of tranquility Minimize

disruption


from the tower will have an anticipated noise level of 65 dB at 50 ft from the

tower. Note that the curve is not linear and an increase to 25 times the sound

rating distance from the tower only reduces the sound level by approximately

29 dB.

To avoid a community noise problem, locate the tower as far from adjacent

property lines as possible. Since HVAC cooling towers have sound levels of 65–

85 þ dBA, the WHO criteria requires that cooling towers not be located within

50–100 ft of property lines without taking additional attenuation measures:

1. Select/approve the quietest cooling tower possible for the application.

Use centrifugal fans or select low rpm propeller fans for the application

and/or take advantage of manufacturer options for reducing tower noise.

2. Orient a quiet side of the equipment toward any sound reflecting wall or

side of a building and toward adjacent property. Locate cooling towers

at least 20 ft away from any sound reflecting surface, no matter how

oriented.

3. Reduce radiated noise with a solid wall sound barrier between the

equipment and the adjacent property. However, take care not to create a

tower performance problem.

4. Only as a last resort should sound attenuators on the tower inlet or

outlet be utilized. These devices are expensive, seriously impact the

performance of the tower (requiring, often, that a larger size be used),

and create significant maintenance problems.

Cooling tower vibration results when fan balance and/or drive alignment is not

correct. This is more often a problem with propeller fans and/or gear drives.

Propeller fan blades, due to their relatively long length (large length-to-width

ratio) tend to flex under air loading. Even if the blades are perfectly balanced by

weight, dynamic balancing is required to eliminate unequal blade movement and

resulting vibration. Shaft alignment into and out of gear drives is also critical to

TABLE 14.4. Effect of Distance on Cooling

Tower Sound Level

Multiple of

sound rating

distance

Approximate

sound reduction

(dB)

5 £ 13

10 £ 20

15 £ 24

20 £ 26

25 £ 29

Chapter 14220

avoid vibration. Also, gear drives must be operated at speeds high enough to

avoid “gear chatter.”

Centrifugal fans, unless they have very long shafts, have less vibration

potential because of their construction. However, many cooling towers using

centrifugal fans will be constructed with two or more fan wheels on a common

shaft, driven by a single motor. To prevent vibration, the shaft must be carefully

aligned and supported by bearing blocks at each end and between each fan wheel.

The shaft must be rigid enough to avoid deformation, and the resulting vibration,

due to the fan wheel(s) weight.

All propeller fan and/or gear drive cooling towers, and towers with multiple

fan wheels on a common shaft, should be equipped with a vibration cutout switch

(see Chap. 12).

14.4. PLUME CONTROL

If HVAC cooling towers are operated during the winter, a significant and

objectionable discharge plume can result. When warm air, at or near saturation

conditions, leaving the cooling tower impacts with the cold ambient air,

condensation occurs, resulting in a super-saturated air stream and visible “fog.”

This fog, trapped in the moving discharge from the cooling tower, is distributed

upward and outward to form the plume.

Uninformed observers often equate visible cooling tower plume with

“smoke,” indicating a fire, or with “pollution.” In either case, 911 phone calls can

result.

Generally, plume problems can be handled with education. People who are

routinely around the tower simply become used to this harmless condition. Fire

and police dispatchers should be informed that this condition is not a problem so

they can handle public alarm accordingly.

In a small number of cases, cooling tower plume can become a more serious

problem. Since it can return to ground level to form a vision hazard or, in very cold

weather, cause ice to formon downwind surfaces and structures. If plume abatement

does become necessary, there are two basic approaches that can be applied:

1. The tower can be equipped with bypass heating as shown in Figure 14.5.

In this configuration, additional ambient airflow is induced into the tower

discharge. Enough heat is added to this induced air to raise its

temperature high enough to allow it to absorbmoremoisture and prevent

condensation until after the discharge plume has dissipated.

2. The tower can be equipped with reheat coils in the discharge air as

shown in Figure 14.6. These coils are sized to provide enough heat to

offset the cooling effect of the ambient air on the plume and, thus,

prevent condensation until after the discharge plume has dissipated.


FIGURE 14.5. Tower arrangement for plume control via bypass air heating.

FIGURE14.6. Tower arrangement for plume control via discharge air reheating.

Chapter 14222

Obviously, the need to introduce heating to the tower is expensive to install and to

operate and should be used only as a last resort.

14.5. FIRE PROTECTION

The National Fire Protection Association (NFPA) Standard 214 defines

“noncombustible” as any material that will not ignite, burn, support combustion,

or release flammable vapors when subjected to fire or heat. When any HVAC

cooling tower is constructed wholly or partly with materials that are not

“noncombustible,” a fire protection (sprinkler) system is required by the NFPA

standard. (See local codes for applicability of NFPA 214.)

Any of four types of sprinkler systems, applied and designed in accordance

with NFPA Std. 13, may be used, as follows:

1. Counterflow towers may be protected with a wet-pipe system, a dry-

pipe system, a preaction system, or a deluge system.

2. Crossflow towers must be protected with a deluge system.

The minimum required rate of water application must be in accordance with

Table 14.5.

When the fan deck is combustible and a deluge system is applied, the

underside of the deck must be protected with a minimum water application rate of

0.15 gpm/sf. With crossflow towers, when wet deck covers are constructed of

combustible materials, protection under these covers, at the same rate as for the

fan deck, is also required.

All sprinkler piping, fittings, hangers, braces, attachment hardware, etc.,

must be galvanized to reduce corrosion. Exposed pipe threads and bolts must also

be protected against corrosion.

Obviously, freeze protection for the sprinkler system is of paramount

importance. While a wet-pipe system is allowed by NFPA, it really has only very

limited geographic application in the United States and is rarely used. The most

common systems applied are the (closed head) dry pipe system, usually applied

to smaller towers or where the water supply may be limited, and the (open head)

deluge system.

TABLE 14.5. Fire Protection Water Flow Rate Requirements

Location Rate (gpm/sf)

Under fan decks (including fan opening), counterflow tower 0.50

Under fan decks (including fan opening), crossflow tower 0.33

Over fill area, crossflow tower 0.50


Overall, the deluge system provides a higher degree of protection when

there are adequate water supplies and this system has the advantage of

minimizing the possibility of failure due to freezing.

Chapter 14224

15

Cooling Tower Operation andMaintenance

15.1. TOWER COMMISSIONING

When the cooling system’s installation has been completed, it is necessary to start

the cooling tower and place it in service. Condenser water system commissioning

can be broken down into three basic elements, with numerous requirements

associated with each element, outlined as follows:

A. Condenser water pump

1. Check pump installation, including mountings, vibration isolators and

connectors, and piping specialties (valves, strainer, pressure gauges,

thermometers, etc.).






B. Cooling tower

1. Clean all tower surfaces, flush and clean the wet deck and basin.

2. Clean the basin strainer.

3. Lubricate fan shaft bearings as required by the manufacturer.

225


5. Test and adjust belt drive (if installed).

a. Adjust belt(s) tension.

b. Check and adjust belt(s) alignment.

6. Test and adjust gear drive (if installed).

a. Fill oil reservoir.

b. Check shaft alignment.

c. Check couplings for bolt tightness, excess play, etc.

7. Turn shaft by hand to make sure fan and drive turn freely.

8. “Bump” motor(s) on and check fan(s) for correct rotation.

9. Run fan(s) for a short periodand check for unusual noises and/or vibration.

Verify that motor amps are in accordance with manufacturer’s data.

10. Confirm that the condenser water piping has been flushed and

chemically cleaned.

11. Fill the basin and piping to the manufacturer’s recommended basin

operating level.

12. Run condenser water pump for a short period.

a. Verify that pump motor amps are within motor nameplate rating.

b. Test the wet deck distribution, fill, and basin for proper water flow.

c. Check basin for vortexing.

13. Test basin freeze protection thermostat and heater.

14. For multi-cell installations:

a. Ensure that automatic isolation valves function properly.

b. Balance condenser water flow to and from each cell.

c. Balance equalizer line to maintain proper basin operating level in

each cell.

15. Place tower fan and condenser water motor starters in “automatic”

position.

C. Controls

1. Check that all temperature sensors are properly installed.

2. Test operation of bypass control valve and adjust for 708 rotation.3. Check controller set point.

4. Test fan control relays for proper functioning.

5. Test VFD (if installed) for proper operation.

6. Test operation of vibration cutout switch.

7. Test operation ofbasin level controls, includinghigh and low level alarms.

8. Place controls in operation.

Chapter 15226

D. Water treatment

1. Check that all pH and/or conductivity probes are properly installed.

2. Test operation of blowdown solenoid valve.

3. Place water treatment equipment in operation.

At this point, the condenser water system should operate and be controlled to

maintain set-point temperature under its automatic controls. However, during the

first 24 hr of operation, all aspects of condenser water pump, cooling tower,

controls, and water treatment should be checked and evaluated frequently.

Unusual noises or vibration, erratic performance or operation, or other problems

(see Sec. 15.3) may require that the system be shutdown and repaired before

being placed into fulltime service.

Once the tower has been in operation for several days and all start-up

requirements have been met, test the tower for excess recirculation. This

condition can be determined by measuring the ambient air wet bulb temperature

and the wet bulb temperature entering the tower. An increase of more than about

18F in the entering wet bulb over the ambient wet bulb indicates that excess

discharge is re-entering the tower. This condition must then be corrected based on

the parameters outlined in Sec. 11.2.

15.2. COOLING TOWER MAINTENANCE

Once the cooling tower system is placed into fulltime service, routine inspection

and maintenance must be done to ensure proper tower operation and to obtain the

expected service life of the equipment. The required maintenance can be broken

into two areas: water treatment and mechanical maintenance.

15.2.1. Water Treatment Management

Water treatment requirements and programs to prevent corrosion, deposition, and

biological fouling in condenser water systems have been addressed in Chapter 13.

However, as a routine matter, the tower owner must ensure that this requirement

is being met effectively by his water treatment contractor. To this end, the

following procedure is recommended:

1. Require and evaluate regular and frequent reports by the water

treatment contractor, first to ensure that regular water treatment is being

done and, second, to “track” the various treatment parameters such as

pH, TDS, chemical types, quantities used, etc.

2. At least twice each year, send a water sample to an independent

laboratory for analysis and compare the results with the most recent

monthly report from the water service contractor.

Cooling Tower Operation and Maintenance 227

3. During shutdown periods, the maintenance staff should inspect the

tower and as much piping as possible for scaling or fouling that is being

inadequately addressed by the water treatment program.

4. Track chiller and tower performance on a routine basis to determine if

the system is remaining free of deposition or fouling. The information

shown in Figure 15.1 should be logged on a regular basis (at least once

per day or once per shift) and, with the data from Figure 7.1, the

relationships between load, temperature difference, and power input

tracked. If the relationships between these values change appreciably,

this could indicate chiller fouling, tower fouling, or other performance

problems.

15.2.2. Mechanical Maintenance

Routine cooling tower mechanical inspection and maintenance falls into two

categories: start-up and operating.

FIGURE 15.1. Cooling tower maintenance log.

Chapter 15228

When the tower is to be started after being shutdown for a lengthy period of

time, it must be thoroughly inspected and repaired, as follows:

1. Check drift eliminators for proper position, being clean, etc.

2. Check fans, bearings, motors, and drives for proper lubrication.

3. Rotate fan shaft(s) by hand to make sure they turn freely.

4. Check fan motors for proper rotation and adjust belt tension for belt

drives.

5. Fill basin with fresh water and check operation of level controller.

6. Start condenser pump and check wet deck for proper distribution.

7. Check fill for fouling and/or clogging and clean or replace if necessary.

8. Check access door gaskets and replace as necessary.

9. Throughly inspect all metal surfaces for corrosion, scale or fouling, or

sludge. Clean as required and any damaged metal should be cleaned

down to bare metal and refinished with a cold zinc coating.

10. Operate tower and look for and repair any water or air leaks from the

basin, casing, or piping.

During the cooling season, regular inspection and maintenance of the tower

is required to ensure proper operation, as follows:

A. Weekly

1. Clean basin strainer.

2. Check blowdown valve and make-up water valves to make sure they

are working properly.

3. Test water and adjust chemical treatment as necessary.

4. Check/fill gear drive oil reservoir.

B. Monthly

1. Clean and flush basin. (This may be required more often for towers

located adjacent to highways, industrial sites, etc. with high particulate

emissions.)

2. Check operating level in basin and adjust as necessary.

3. Check water distribution system and sprays.

4. Check drift eliminators for proper position.

5. Check belts or gearbox and adjust as necessary.

6. Check fans, inlet screens, and louvers for dirt and debris. Clean as

necessary.

7. Check keys and set screws.

C. Regularly

1. Lubricate motor in accordance with MFGR’s instructions.

2. Lubricate fan shaft bearings every 1000 hr or 3 months.


3. Check gear drive in accordance with MFGR’s instructions.

4. Blowdown condenser water pump strainer.

D. Yearly

1. Clean and touch-up paint or other protective finish as necessary

(including grillage).

2. Dismantle and clean condenser water pump strainer.

15.2.3. Induction/Venturi Tower Maintenance

Due to its lack of fan, motor, and drive assembly, routine maintenance for the

induction draft or venturi cooling tower is somewhat simpler than for a

mechanical draft cooling tower. Basically, except for the fan elements, routine

maintenance should include the same elements and schedule outlined in

Sec. 15.2.2.

There are several special maintenance aspects of this type of tower that

should be addressed on a monthly basis as follows:

1. Strainers: There are two strainers that must be removed and cleaned by

high pressure water spray. The sump strainers are flat screens that

require the basin to be drained before removal. A final strainer is

located in a vertical cylindrical enclosure on the inlet side of the tower.

This strainer is removed by removing the bolted cover plate.

2. Spray nozzles: Inspect the spray nozzles while the tower is in operation.

If cleaning is necessary (indicted by poor or erratic spray), it can

usually be accomplished while the tower is in operation by scrubbing

the nozzle with a nonmetallic brush and using a nonmetallic pick. If this

cleaning is unsatisfactory, shutdown the tower, remove the nozzle, and

clean thoroughly in the shop.

3. Inlet louvers: The inlet louvers are much more closely spaced than on

other types of towers and must be kept clean. They are removable to

facilitate cleaning (and to provide access to the spray nozzles).

15.2.4. Heat Exchanger Maintenance

When a waterside economizer cycle is utilized, the heat exchangers applied must

also be maintained on a regular basis, as follows:

1. Plate and frame heat exchanger: When the economizer cycle is in use,

perform the following routine maintenance:

a. At least one manufacturer recommends that during the summer

months when the heat exchanger is not in use, the plate pack should

be loosened until all gasket compression is removed.

Chapter 15230

b. At the beginning of the mechanical cooling season, when the

economizer cycle is no longer used, dismantle and clean the heat

transfer plates. High pressure water can be used, avoiding the

gaskets, along with brushing with a fiber bristle brush or a brush of

the same alloy as the plates. Replace gaskets as necessary.

c. Make periodic inspection of control valves to ensure proper

operation.

2. Shell and tube heat exchanger: When the economizer cycle is in use,

perform the following routine maintenance:

a. Monthly, flush the shell and tube bundle and inspect the interior

and exterior condition of the tubes. (Tubes can be cleaned with a

10% muriatic acid solution allowed to stand for 2 hr.) Replace

gaskets when replacing heads.

b. Make periodic inspection of control valves to ensure proper

operation.

15.3. TOWER PERFORMANCETROUBLESHOOTING

Selected correctly, installed correctly, and properly maintained, the HVAC

cooling tower should perform as required for many years. However, the tower

may not perform as anticipated due to numerous problems that must be

investigated and evaluated in the field.

Typical HVAC cooling tower performance problems fall into three major

categories:

1. Selection problems

2. Installation problems

3. Maintenance problems

15.3.1. Selection Problems

Performance problems related to the tower selection refers to inadequate tower

performance, because one or more of the tower performance requirements was

incorrectly or inadequately specified.

From Chap. 9, two factors were defined as establishing the HVAC cooling

tower performance:

1. The cooling load imposed on the tower, which is defined by the

condenser water flow rate and the selected range.

2. The approach, which is dictated by the design outdoor wet bulb

temperature and the required condenser water supply temperature.


Of these variables, the most common errors are made in defining the cooling load

and in selecting the design outdoor wet bulb temperature. If either or both of these

variables is under estimated, the cooling tower selected will simply be too small.

Selection errors are usually identified by performance problems when at or

near design conditions, and the design condenser water supply temperature

cannot be achieved. Measuring the outdoor wet bulb temperature under these

conditions will demonstrate whether the tower selection was based on too low of

a design wet bulb temperature. The imposed cooling load can be computed based

on the design flow rate and the actual entering and leaving condenser water

temperature and this value compared to the design requirement.

Unfortunately, there is relatively little that can be done to correct selection

problems short of adding additional cooling tower capacity since the tower

characteristic is too small for the performance required.

15.3.2. Installation Problems

There are a host of potential installation problems that can adversely affect

cooling tower performance:

1. Excess recirculation may exist because the tower is not installed with

sufficient clearance around it for free airflow. This condition can be

determined by measuring the ambient air wet bulb temperature and the

wet bulb temperature entering the tower. An increase of more than

about 18F in the entering wet bulb over the ambient wet bulb indicates

that tower discharge is re-entering the tower.

2. There is inadequate airflow through the tower because the fan speed is

not correctly adjusted. This problem can be solved by the adjustment of

the fan drive for the proper fan speed.

3. There is inadequate water flow through the tower because the pump is

incorrectly sized or installed incorrectly. Correcting the pump usually

corrects the flow problem.

4. The make-up water system is inadequately sized and the tower runs

dry. The make-up water supply must be sized in accordance with

Sec. 11.5.2.

5. Bypass piping is incorrectly installed and the chiller will not start on

cold mornings. This problem is corrected by eliminating the potential

air gap in the pump suction as discussed in Sec. 12.1.

6. With two-speed fan systems, time delay is not included in the control

scheme and the drive is damaged. See Sec. 12.2.2.

7. For multi-cell tower installations, the equalizer line may be

inadequately sized if one basin floods while another runs dry. This

problem is usually corrected by installing an equalizer line sized in

accordance with Table 11.3.

Chapter 15232

8. As obvious as it may sound, some cooling towers do not get installed

with the basin level. Thus, the basin operating depth varies from one

end to another (or even worse, from corner to corner). Obviously, basin

level control under these conditions can be difficult and the only real

solution is to disconnect all of the piping, level the tower, and reinstall

the piping.

9. As discussed in Chap. 11, if any of the condenser water supply piping

between the tower basin and the pump suction is installed at an

elevation higher than the tower operating level, the pump will loss

prime due to air entrainment. The only fix for this problem is lower the

piping or raise the tower.

15.3.3. Maintenance Problems

Proper water treatment and regular mechanical maintenance of the HVAC

cooling tower is required to keep it performing as needed (see Sec. 15.2).

15.4. ENHANCING TOWER PERFORMANCE

When the load on a cooling system increases, the load imposed on the condenser

water increases in direct proportion. However, if the load increases to a point

where the cooling tower capacity is exceeded, the tower must be modified, if

possible, to enhance its performance.

Increased load imposed on an HVAC cooling tower is represented by

increased condenser water flow rate and/or increased range. Increased tower

performance is achieved by increasing the airflow through the tower and/or

increasing the tower heat exchange (fill) area. Thus, the first step in the analysis is

to determine a new tower characteristic and from that determine the required

additional heat transfer area and airflow rate required. The manufacturer can then

determine if increasing airflow and amount of fill is possible. If not, the next step

is to add new cooling tower to operate in parallel with the existing tower.

Induced draft towers have, usually, more potential for increasing airflow

than do forced draft towers. The induced draft fan speed and/or blade pitch may

be changed to produce more airflow. The fan wheel can be replaced with one with

more blades. As a last resort, the entire fan assembly can be replaced with a larger

one. In each of these cases, the fan motor size will increase as the cube of the

increased airflow and must be replaced, also requiring wiring changes to support

the increased motor horsepower.

Fill can be replaced with a different type that is more efficient. Bar splash

fill can be replaced with formed PVC fill that is 15–20% more effective. It may

even be possible to replace splash fill with film fill, improving heat transfer by as

much as 25%. Existing film fill can be replaced with more efficient film fill, but at

the expense of increased air pressure drop and fouling potential.


The tower configuration itself can be changed to improve performance. For

example, a forced draft tower can be reconfigured as an induced draft tower or the

small fans of a forced draft tower can be replaced with ducted air delivered from

much larger fans. Usually, however, this type of modification, except in very

large towers, is not cost effective, and the inadequate tower should simply be

replaced with a new one.

15.5. COOLING TOWERS IN FREEZING CLIMATES15.5.1. Winter Tower Operation

Winter tower operation may be dictated by either the need to provide cooling

when systems have no economizer cycle or to serve in a waterside economizer

application. Typically, the freezing problem is not related to the main condenser

water flow stream, or even the water in the tower basin, but to ice formation from

stray water droplets and/or saturated air impinging on cold surfaces.

Since forced draft cooling towers are far more susceptible to icing

problems and potential mechanical damage, induced draft towers are usually

used where winter operation is required and, therefore, this discussion is limited

to that tower configuration.

Ice, in small quantities, will often form on tower louvers, structure, and the

entering edge of the tower fill and is rarely a cause for concern. However, if the

ice thickness grows, it can extend across and into the fill, effectively blocking

airflow and, by its weight, cause structural damage to the tower. Thus, it is

necessary for the tower operators to monitor ice formation and to take routine

steps to control ice formation:

1. Never operate the cooling tower without a heat load. Without a heat

load being imposed, the water temperature will continue to fall until (a)

the ambient wet bulb temperature is reached or (b) ice forms,

whichever occurs first.

2. Do not modulate water flow through the tower. Reducing the water flow

will provide low flow regions in the fill that will promote icing. If flow

control to prevent the water temperature from going too low is required,

use the bypass valve as a two-position device rather than a modulating

device. . .send all of the water through the tower, then bypass all of the

water.

3. Control airflow through the tower to control condenser water supply

temperature above freezing. As discussed in Chap. 12, tower capacity

control is routinely done by controlling the amount of air through the

tower. Two-speed or pony motor arrangements are much better than

single-speed fan cycling and should always be used for towers

requiring winter operation.

Chapter 15234

4. De-ice the tower by airflow reversal. If, despite best efforts to prevent

it, a significant amount of ice does buildup on the tower, de-icing can be

accomplished by reversing the fan rotation and, thus, the airflow

direction though the tower. This warm air is circulated back through the

fill and louvers to melt the ice that has formed on these surfaces. Fan

reversal can be accomplished by installing a forward-off-reverse switch

in the fan circuit or, for three phase motors, simply switching two of the

motor wires manually. To prevent damage to the fan and tower, fan

reversal should follow this procedure:

a. Turn the fan off for 2–3min.

b. Run the fan in reverse for not more than 15–20min at one time.

c. If the tower has a two-speed or pony motor, run the fan in reverse at

low speed.

d. If the tower consists of multiple cells, de-ice each cell individually

while the adjacent cells are shut down.

e. After reversal, turn the fan off for 5–10min before restarting in the

correct direction.

f. Check fan and drive for proper operation, vibration, noise, etc.

5. Never use antifreeze solutions in condenser water. The use of

antifreeze chemicals in the water changes the evaporation rate of the

water and reduces the efficiency of the tower and, upon evaporation,

pollutes the area around the tower.

6. During cold weather, monitor the tower carefully. Never allow the

tower to operate unattended.

15.5.2. Winter Tower Shutdown

If the tower is not required to operate during the winter, then it can be shutdown

for the heating season. To protect the tower from freezing and to minimize

adverse affects to the system, the following guidelines are recommended for

winter shutdown:

1. The tower wet deck and basin must be drained to prevent freezing.

2. Condenser water piping should be drained only to where it enters the

building or is otherwise protected from freezing. Studies have shown

that piping systems that are drained tend to have significantly increased

corrosion rates due to exposure to air and the resulting oxidation. Thus,

drain the minimum amount of piping possible needed to prevent

freezing and keep the rest of the system flooded.

3. Drain exposed make-up water piping. (This piping is usually copper or

plastic and corrosion is not a problem.)


An alternative to the manual shutdown and draining outlined above is to design

the tower system with a remote sump within the building so that the tower

operates effectively “dry.” When the tower is shutdown, all water drains to the

indoor sump and no additional freeze protection is required. A typical tower and

remote sump arrangement is illustrated in Figure 15.2.

Corrosion may occur and fouling and/or scale may form in the shutdown

condenser water system. Corrosion shows up during the following cooling season

when iron rust and scale breaks loose and begins to plug wet deck nozzles,

strainers, and/or condenser tubes. Scale forms when wetted pipe dries and begins

to rust. Other fouling is formed by the dissolved solids are left behind as the water

evaporates. To mitigate these problems, the following shutdown procedures are

recommended:

1. As the cooling season ends, decrease the cycles of concentration to

approximately 5–6 to reduce the level of dissolved solids in the

water.

FIGURE 15.2. Indoor sump schematic.

Chapter 15236

2. Before shutdown, add a nonoxidizing biocide to maximize biological

“kill,” since stagnant water left in the system will enhance the

environment for biological growth.

3. Add additional dispersant to loosen and penetrate existing fouling.

4. To minimize corrosion during the shutdown period, circulate a film

type corrosion inhibitor for 3–5 days before the system is shut down

and drained, but only after sufficient time as elapsed for the dispersant

to have been effective.

5. All water treatment systems should be shut down, cleaned, and flushed

with fresh water in preparation for the next cooling system.

Conductivity and pH probes should be removed, cleaned, and stored

in a cool, dry area.


16

Buying a Cooling Tower

16.1. DEFINING TOWER PERFORMANCEREQUIREMENTS

The first step in purchasing a new or replacement HVAC cooling tower is to

define the performance or capacity for it. To do this, the following four

parameters must be specified:

1. Condenser water flow rate, gpm

2. Entering (return) condenser water temperature, 8F3. Leaving (supply) condenser water temperature, 8F4. Entering (ambient) wet bulb temperature, 8F

The difference between the entering and leaving condenser water

temperatures defines the required range, while the difference between the

leaving condenser water temperature and the entering wet bulb temperature

defines the required approach. With range and flow rate, the total Btu/h of

heat rejection capacity is defined. With this data, any manufacturer can select

a cooling tower.

The next step is to define one or more tower configurations that would be

acceptable, as follows:

238

Table 16.1 summarizes the typical types of construction available for various

draft and flow arrangements. As discussed in earlier chapters, wood and

concrete/masonry towers are used infrequently in HVAC applications unless the

tower requirement is very large (greater than 2000 ton) or there is a specific

aesthetic element that must be addressed.

When one of the acceptable alternatives is a galvanized tower, it is

recommended that the basin and, for crossflow towers, the wet deck be

constructed of stainless steel. This option is offered by all of the tower

manufacturers and will increase tower costs by 20–40%. But, as shown in

Table 16.2, the typical cooling tower life is extended by at least 33%, making it

an excellent investment.

DRAFT/FAN A—Forced/propeller fan

B—Forced/centrifugal fan

C—Induced/propeller fan

Flow arrangement 1—Crossflow

2—Counter flow

Assembly Factory assembled

Field erected

Construction Wood

Galvanized steel

Galvanized steel with stainless steel basin (and wet deck, for a

crossflow tower)

All stainless steel

Fiberglass reinforced plastic (or equivalent)

Concrete/masonry

TABLE 16.1. Cooling Tower Construction Options

Typical construction availability

for tower type

Construction A1/B1 C1 A2/B2 C2

Wood X

Galvanized steel X X X

Galvanized steel with stainless steel basin

(and wet deck, as applicable)

X X X

All stainless steel X X X

FRP X X X

Concrete/masonry X X X

Buying A Cooling Tower 239

Factory-assembled towers are much easier and faster to install and tend to

be less expensive. Thus, field-erected towers are rarely used for HVAC

applications except, again, for very large loads (.2,000 tons).

16.2. CTI RATINGS AND PERFORMANCEGUARANTEES

Except in very special circumstances or for large field-erected towers, HVAC

cooling towers should be specified tohave their performance certifiedby theCooling

Technology Institute (CTI) in accordance with CTI Standard 201. This certification

amounts to a third party guarantee that the towerwill perform as advertised provided

the tower is installed correctly and the condenser water is “clean.”

CTI certification eliminates the need and expense of field-testing a cooling

tower to confirm that it performs as specified. Table 16.3 lists the towers that are

CTI certified. However, new towers are added to CTI’s list from time to time, so

the list should be checked (www.cti.org) to confirm any proposed cooling tower’s

certification.

16.3. ECONOMIC EVALUATION OF ALTERNATIVECOOLING TOWER SYSTEMS

Life cycle costing is the best way to compare alternative cooling tower selections.

It provides for a consistent method of analyzing the economic aspects of each

tower and allows realistic comparison so that the most cost-effective, e.g., least

life cycle cost, tower can be selected. The process is basically simple and requires

only that each potential tower be evaluated using the same criteria. The result is

an “apples-to-apples” comparison, not “fruit salad.”

The computations to determine life cycle cost utilizing the total owning and

operating cost methodology are simple. However, the methodology’s accuracy

TABLE 16.2. Estimated Economic Life

Equipment Life (years)

Pump, base-mounted 20

Pump, line-mounted 10

Cooling tower, galv. steel 15

Cooling tower, FRP 20

Cooling tower, galv. steel w/s.s. basin and wet deck 22

Cooling tower, wood 25

Cooling tower, all stainless steel or masonry 35

Chapter 16240

depends wholly on the accuracy of the data utilized. Anyone can calculate

“garbage” life cycle costs simply because they use data and/or assumptions that

are “garbage.” Two different individuals, faced with the same evaluation may

compute wildly different life cycle costs because they use significantly different

data and/or assumptions in their computations.

The following subsections define the basic elements that make up the life

cycle cost.

TABLE 16.3. CTI Certified Cooling Towers (May, 2002)

Manufacturer Cooling tower models Configurationa

Baltimore AirCoil Company, Inc. FXT line A1

V VF1 B2

V VFL B2

V VTL B2

V VTO B2

V VT1 B2

Series 1500 C1

Series 3000A C1

Series 1500 FXV C1

Ceramic Cooling Tower Company,

Inc. (A division of

Baltimore AirCoil Company, Inc.)

Ultralite PSC-XF line C1

Ultralite XL100 line C1

Evapco, Inc. AT series C2

SST series C2

UBT series C2

USS series C2

Fabrica Mexicana De Torres, SA GHR line C2

Reymsa Cooling Towers

Integrated Tower Systems, Inc. ITF series

Kyung In Machinery Company, Ltd EX line C1

Liang Chi Industry Company, Ltd LRC line

Marley Cooling Tower Company NC series C1

Quadraflow series C1

Sigma Line C1

AV series C1

Aquatower 3900 & 4900 series C1

Primus series C1

Tower Tech, Inc. TTGE line A2

a See Sec. 16.1 for configuration codes.


16.3.1. First Costs

The initial capital costs associated with each potential cooling tower are all of the

costs that would be incurred in the design and construction of that tower and the

associated condenser water pump. Equipment costs can be obtained directly from

the prospective equipment vendors. Installation cost estimates can be obtained

from local contractors or, lacking that, from cost data published by R. S. Means

Co., Inc. (Construction Plaza, 63 Smiths Lane, Kingston, MA 02364-0800,

781/585-7880 or 800/334-3509).

The construction cost estimate must include the following, in addition to

the cost of the tower itself:

1. Tower dunnage and grillage

2. Rigging

3. Demolition

4. Electrical power

5. Controls

6. Contractor Overhead (Insurance, bonds, taxes, and general office

operations, special conditions), typically 15–20%

7. Contractor Profit, typically 5–20%

Other costs that may be included in the capital requirement are design fees,

which may increase or decrease as a function of the selected alternative; special

consultants’ fees; special testing, etc. Also, condenser water, make-up water,

and/or drainage piping may change configuration and cost between alternative

towers.

Unless the cost of the tower is being met from operating revenues, at least a

portion of the capital expense will be met with borrowed funds. The use of this

money has a cost in the form of the applied interest rate and information about

the amount of borrowed funds, the applied interest rate, and the period of the loan

must be determined for the analysis. Then, the total capital cost, including both

principal and interest can be computed using Eq. (8.1).

16.3.2. Annual Recurring Costs

Once the tower system is placed into operation, two annual recurring costs must

be met each year of its economic life: energy costs and maintenance costs.

The economic life for an alternative is the time frame within which it

provides a positive benefit to the owner. Thus, when it costs more to operate and

maintain a piece of equipment than it would to replace it, the economic life has

ended. Typically, the economic life (sometimes called “service life”) is the period

over which the equipment is expected to last physically. The typical economic

life for the cooling towers and pumps is provided in Table 16.2.

Chapter 16242

The computation of energy cost requires that two quantities be known: (1)

the amount of electrical energy consumed by the tower and the associated

condenser water pump and (2) the unit cost or rate schedule for that energy. The

second quantity is relatively easy to determine by contacting the utilities serving

the site or, for some campus facilities, obtaining the cost for steam, power, chilled

water, etc., that may be furnished from a central source.

To accurately evaluate and compare alternative cooling towers, it is

necessary to also consider the condenser water pump. Since different towers may

impose different pressure losses on the pump, this energy and cost component

must be incorporated into the analysis.

To determine the total energy use by the condenser water pump and cooling

tower, the first step is develop a “load profile” for the condenser water system in

terms of system load (tons) as a function of outdoor temperature, usually for each

five degree “bin” of outdoor temperature. At each bin, the power input to the

condenser water pump and cooling tower must be determined in terms of kW and

then multiplied by the hours of occurrence of that bin to yield energy

consumption in terms of kWh. Adding the energy consumption in all of the bins

yields the total annual energy consumption. Figure 16.1 provides a form that can

be used for this calculation.

With this form, for single speed motors, the “percent load” is 100% any

time the fan is on. For 2-speed or pony motor arrangements, the full speed load is

100%, while the half speed load is approximately 30%. When a VFD is used, the

percent load can be estimated from Figure 12.6.

Cooling system load profiles fall into three basic forms:

1. Where airside economizer systems are used or the system is located in a

cold climate and is not operated in the winter, the load can be prorated

between the design summer temperature and about 558F. In each of

these bins, since some cooling is required, the condenser water pump

must run and, therefore, its load is 100% in each bin. The tower fan,

with cycling capacity control, will operate a percentage of the time

essentially equal to the percent load imposed on it.

2. When the cooling system operates on a year around basis, the cooling

load will be prorated to all bins of temperature.

3. When a waterside economizer is in use, the winter use of the condenser

water pump and cooling tower must be estimated. Again, if there is any

cooling load imposed in a bin, the condenser water pump power

requirement is 100% in that bin. The tower fan use can be prorated in

direct proportion to the imposed cooling load in the winter.

The energy cost, then, is computed by multiplying the electrical energy

consumption by the unit cost for electricity.


Annual recurring maintenance cost is a very difficult element to estimate.

Lacking other information, the annual routine maintenance cost associated with

pumps and cooling towers can be estimated as a percentage of the initial

equipment cost, as follows:

Galvanized steel or wood cooling tower 3%

Galvanized steel tower with stainless wet deck and basin or Fiberglass

reinforced plastic (FRP) tower 2%

Stainless steel or masonry tower 1%

These costs do not include the cost of make-up water or the water treatment

program. Since the amount of make-up water and water treatment chemical

consumption is a function of (a) cooling tower loads, (b) the water treatment

program required, and (c) the water flow rate, it is essentially independent of the

FIGURE 16.1. Condenser water system energy consumption calculation form.

Chapter 16244

tower itself. Therefore, this element can be ignored when comparing alternative

cooling towers unless there is a significant difference in drift losses between

alternative towers. (The one exception to this is the additional water treatment

required for wooden towers and this additional cost should be included in the

analysis.)

16.3.3. Nonrecurring Repair and Replacement Costs

Nonrecurring costs represent repair and/or replacement costs that occur at

intervals longer than 1 year. For example, a galvanized wet deck and basin in a

crossflow induced draft tower may require significant repair (or even

replacement) after 10 years of service. These costs must be determined and the

year of their occurrence estimated.

16.3.4. Total Owning and OperatingCost Comparison

The total owning and operating cost for a cooling tower and its associated

condenser water pump, over the system economic life, can be computed using

Eq. (16.1).

Life Cycle Cost ¼ C þ ðSum of Repair=Replacement CostsÞþ ½ðEconomic LifeÞ £ ðAnnual Energy Cost

þ Annual Maintenance CostÞ� ð16:1ÞC is the total capital cost computed from Eq. (8.1).

16.4. PROCUREMENT SPECIFICATIONS

Recommended specifications for both induced draft and forced draft cooling

towers are provided in Appendix B. These specifications are based on factory-

assembled galvanized steel towers with stainless steel basins and, for crossflow

towers, stainless steel wet decks. Each specification includes both crossflow and

counterflow configurations and, therefore, must be edited carefully if only one of

these configurations is acceptable.

16.5. WATER TREATMENT PROGRAMCONTRACTING

The following are the recommended minimum standards that should be applied

to the selection of a water treatment service company:


1. To be considered, a water treatment service company should be an

established company with full service capabilities. The company

should have been in business at least five years, have corporate staff

with sufficient expertise and experience to competently address all

aspects of water related issues, and be capable of providing a

reasonable list of references for whom they have provided service for a

minimum of two years. Call and verify the listed references.

2. The local service personnel should have sufficient expertise and

experience to competently address all aspects of water related issues.

After all, it is the local service technician who will actually provide the

day-to-day service, not a “water expert” at the corporate office. Ask for

a detailed resume for the service technician. Further, check into the

company’s service technician turnover rate. If the local technician is

constantly changing, the quality of service provided by this company

will be uneven.

3. The proposed approach to water treatment should incorporate proven

technology. This does not mean totally excluding “new” technologies,

products, or methods, but it does mean that the water treatment service

company must demonstrate that their proposed technology has been

successfully applied at other locations with similar water treatment

conditions and needs. Ask for a list of these locations and call them.

4. Ensure the program performance and its cost. Even with the most

reputable water service companies, there are many instances where the

water treatment program performance and/or cost did not meet

promises. Therefore, develop detailed and rigid performance and cost

standards and include them in the contract with the service company,

including penalties if the standards are not met.

As a basis of comparing alternative water treatment programs by different service

companies, prepare a detailed “request for proposals (RFP)” that defines

requirements and standards and send it to several service companies who

potentially meet the standards outlined above. Thus, there will be several

different proposals that will have a common basis for comparison. For larger

systems, it may be necessary to retain an independent consultant to develop the

RFP and/or evaluate the vendor proposals. Costs incurred for the consultant will

be more than offset by the proper selection of a cost-effective water treatment

program.

Initial water service treatment contracts should be for a two-year period.

This gives the service company ample opportunity to address all of the problems

that may be in the system and meet the cost and performance goals that have been

established. After this initial period, contracts should be for one year.

Chapter 16246

17

In-Situ Tower PerformanceTesting

For large towers or for towers with special requirements that are not CTI certified,

in-situ testing is the only way to guarantee that the tower will perform as required.

This is an expensive and time-consuming process that should be undertaken only

after due consideration.

17.1. WHY IN-SITU TESTING?

Thermal performance testing of an operating cooling tower is a complicated and

expensive undertaking. Costs to test even a medium-sized tower can run well into

five figures.

These tests are normally conducted for one of two purposes:

1. Acceptance tests may be required to demonstrate that the installed

cooling tower meets the performance standards that were specified. For

the majority of HVAC applications, tower performance for many

brands and types of cooling towers are certified by the CTI, and it is

more cost-effective to simply select a certified tower and, thus, avoid

the cost of acceptance testing entirely. In some cases, such as for very

large towers or field-erected towers, an acceptance test may be

necessary.

247

2. Performance testing to evaluate a cooling tower’s performance relative

to correcting problems and/or changing the tower’s thermal

requirements. In-situ performance testing yields actual tower

performance data that can be used to indicate required tower

modifications or improvements.

17.2. TESTING CRITERIA AND METHODS

The field testing of cooling tower performancemust be done in accordancewithCTI

Standard ATC-105 or Standard PTC-23 issued by The American Society of

Mechanical Engineers (ASME). Testing is typically done by third party agencies

and it is recommended that the selected agency be an agency that is licensed by CTI

Commonly, thermal performance tests are referred to as either Class A or B

tests. The Class A test is one conducted by using mercury-in-glass thermometers

and grade level psychrometers, while the Class B test uses a data acquisition

system with psychrometers arranged in an array over the entire air inlet face of

the tower. This does not mean that a data acquisition system cannot be used in

conjunction with grade level psychrometers. It is common to call the air

temperature measurement device used in cooling tower tests a psychrometer.

This is technically incorrect as a psychrometer measures the wet bulb (WB) and

dry bulb temperatures while those instruments widely used today measure only a

single temperature. Depending on how the “cooling tower” psychrometers are

assembled, they may be used to measure either a wet or dry bulb temperature.

It is also very important to recognize the difference between an ambient

and entering WB test. Both ASME and CTI recommend that towers be sized and

tested based on entering WB temperatures. This consideration can affect the size

of the tower selected and the results of thermal tests. An ambient WB is defined as

the temperature of the air mass entering the tower less any influence by

recirculation. For an ambient WB test, at least three WB instruments are located

50–100 feet upwind of the tower.

It is also necessary to measure (or otherwise to account for) the temperature

and quantity of any other air streams (interference) entering the tower, other than its

own recirculation. This interference can come from any other source including other

cooling towers. This can be very difficult, if not impossible to do accurately. The

enteringWB temperature attempts to measure the average temperature of all the air

entering the tower regardless of its source.While this is easier than trying to separate

the influenceof several airmasses, it still requires a careful analysis by the test staff to

ensure that the number of instruments and their locations are adequate.

Most testing today is the Class B test, using data acquisition systems to

measure temperatures. With this test method, the first step is to inspect the tower

to ensure it is ready for the test and to identify points of measurement for the

various parameters. While the test agencies may consider condition of the tower,

Chapter 17248

it is not their obligation to clean, balance, or otherwise adjust the tower for the

test—this is the owner or installer’s responsibility.

In the case of an acceptance test, the installer and/or manufacturer will

normally be much more thorough in this area to ensure the tower’s full potential

is measured.

Once all parties are satisfied that the tower is ready for testing, instruments

are deployed as illustrated in Figure 17.1 and the testing begins. For typical large

HVAC cooling towers, the testing process requires one to two days. However,

weather and operating conditions can sometimes increase this time.

To begin the testing process, the test technicians begin taking data. Usually,

the thermal data are started and monitored for a brief period. If any problems with

instrumentation or conditions are noted, efforts will be made to correct them.

Once any corrections are completed, the test begins and must last at least 1 hr

after steady-state conditions have been achieved.

During the test, the technicians will monitor the system temperatures and

measure the water flow rate and fan power. The two test standards offer

recommendations on deviation from design conditions for the test parameters, as

given in Table 17.1.

While it is preferable to comply with all these limitations, it is not always

possible. CTI licensed testing agencies report deviations from recommended

parameters in 70–75% of all tests. Recognizing this, the standards allow for

deviation provided all parties agree. If at any time during the process, it is

determined a parameter is outside the recommended limitations, all parties must

review the situation and reach a unanimous solution. This can result in data being

discarded and restarts required.

Two parameters, the limits on oil, tar, or fatty substances or the total

dissolved solids in the condenser water are not routinely checked during a tower

test. However, if the tower fails the test, and any party thinks that these agents are

present and could have contributed to the failure, these parameters are measured.

To measure the water flow rate, a pitot tube traverse of the piping carrying

water to the tower is the preferred method. A Wattmeter is used to measure fan

input power on mechanical draft tower systems up to 600V. Temperatures are

measured with thermometers, RTD’s, or thermistors.

The test parameters must be measured at regular intervals during the test, as

given in Table 17.2.

In addition, any other factor affecting the tower’s operation or the data

taken must be recorded. These other factors may include pump discharge

pressure, make-up flow and temperature, blowdown flow and temperature,

auxiliary streams entering the collection basin, etc.

The hot water temperature is normally taken in the wet deck or in a tap in

the piping carrying water to the tower. The cold water temperature is normally

taken at taps on the discharge side of the pumps. In most cases, installations have

In-Situ Tower Performance Testing 249

pressure gauges at this location and these gauges can be replaced with flowing

wells for temperature measurement. If this is not possible, separate taps are

required.

The CTI and ASME standards have defined the instrumentation and

procedures very clearly. Unfortunately, the many installation variations and test

circumstances provide multiple obstacles. This can create serious problems for

the technicians and increased uncertainty of results. The straightforward process

alone does not protect against completely meaningless results. For this reason,

FIGURE 17.1. Test instrument locations for in-situ performance testing.

Chapter 17250

the CTI carefully tests those individuals licensed by CTI to lead tests and inspects

and approves their test equipment. Manufacturers also have highly skilled and

trained staff to participate in the testing process, particularly for an acceptance

test, to help ensure the products are properly evaluated.

Evaluation of the data collected during the test must follow the

requirements outlined by the test standard that is being used, either CTI or

ASME.

17.3. TOWER INSTALLATION REQUIREMENTSFOR TESTING

For a thermal performance test, especially an acceptance test, there are certain

site requirements that need to be met by the customer or his representative. The

TABLE 17.1. Recommended Maximum Parameter Variations from Design

Parameters CTI standard ASME standard

Condenser water flow rate ^10% ^10%

Design heat load Not specified ^20%

Range ^20% ^20%

Fan motor power ^10% ^15%

WB temperature ^158F þ58F/ 2 158FWind velocity (max) 10mph with 1min gusts

not to exceed 15mph

10mph with 1min gusts

not to exceed 15mph

Barometric pressure ^1.000 Hg Not specified

Oil, tar, or fatty substances

in water

10 ppm 10 ppm

TDS in water The greater of 5000 ppm

or 1.1 times design

concentration

10% above design

TABLE 17.2. Tower Test Parameter Recording

Requirements

Parameters Test frequency (readings/hr)

Water flow rate 3

Hot water temperature 12

Cold water temperature 12

Wet bulb temperature 36 per station

Fan power 4 (or monitor continuously)

Wind speed 12


following list includes the common considerations for a standard tower

installation operating on a closed-loop system:

1. Pitot tube taps must be installed in the pipe(s) delivering water to the

cooling tower.

2. Sensor taps must be installed for the measurement of water temperatures.

Hot water temperature can normally bemeasured in the distribution basin

of crossflow towers. Often, the pitot taps can also be a measurement point

for hot water temperature. If site-specific circumstances make neither of

these options acceptable, special taps will be required. The cold water

temperature is normally accomplished at the discharge of the circulating

water pumps. The most common location is at the pressure gauge tap

present onmost systems. If this is not available or applicable on a system,

special taps or another solution must be identified. Measurement in a

tower basin is not acceptable.Measurement of coldwater temperature in a

flume or channel can sometimes be accomplished with acceptable

accuracy, but specifics should be reviewed.

3. At the time of the test, safe access to any elevated points of

measurement must be provided. All access must conform to safe work

practices, OSHA requirements, and any local requirements.

4. Power for test instruments (typically 120/1/60) must be available

adjacent to the tower. On very large towers, multiple sources around the

tower may be necessary.

The tower must be prepared for testing before the test technicians arrive:

1. The tower must be clean. The wet deck must not have damaged,

missing, or plugged nozzles or orifices and be balanced as well as the

design allows. The air inlet should be cleared of any blockage. If the

tower has louvers, they should be in the normal design position, if

adjustable. The eliminators should be free of foreign matter. Fan

discharge should be clear and unobstructed.

2. Water flow and heat load to the tower, or representative cells, should be

as close to design as the system will permit. If the test standard

recommended limitations cannot be met, all parties should review the

situation to agree on the deviation or delay/cancel the test.

3. Any water bypass should be closed and inspected to ensure there is no

leakage.

4. Any source of air leakage such as access doors, mechanical equipment

supports, or holes in the casing or fan cylinders must be closed/blocked.

5. All fans must operating at full speed and cannot cycle during the test

period. In the case of tower fans operating with VFD’s, they should be

placed in by-pass mode.

Chapter 17252

6. The owner or his representative should designate a coordinator

qualified to integrate the testing activity and the normal process

operation being served by the cooling tower.

7. The owner or his representative should have an electrician or qualified

operator available to assist in the measurement of fan power.

All parties to the test must be advised in advance of any special safety issues

required at the site.


Appendices

APPENDIX ADESIGN AMBIENT WET BULB

TEMPERATURESa

TABLE A.1

254

State City WB Temp (8F)

Alabama Birmingham 78

Mobile 77

Ozark 81

Alaska Anchorage 60

Arizona Flagstaff 61

Phoenix 78

Tucson 72

Arkansas Little Rock 80

California Los Angeles 70

Sacramento 71

San Francisco 64

Colorado Denver 65

Florida Jacksonville 79

All others 80

Georgia Atlanta 77

Augusta 79

Savannah 79

Illinois Chicago 77

Springfield 79

Indiana Indianapolis 78

Terre Haute 80

Iowa Des Moines 78

Kansas Dodge City 74

Topeka 79

Kentucky Lexington 77

Louisville 78

Paducah 80

Louisiana Baton Rouge 80

New Orleans 81

Maine Augusta 73

Portland 74

Maryland Baltimore 78

Massachusetts Boston 75

Michigan Detroit 76

Saginaw 77

Traverse City 74

Minnesota International Falls 72

Minneapolis/St. Paul 76

Mississippi Biloxi 81

Jackson 80

Missouri Columbia 78

Kansas City 78

St. Louis 79

Montana Butte 60

Missoula 65

Nebraska Omaha 78

Nevada Las Vegas 71

Reno 63

New Hampshire Concord 74

New Jersey Atlantic City 77

Newark 77

New Mexico Albuquerque 65

Roswell 70

New York Albany 74

Buffalo 74

New York City (JFK) 76

North Carolina Asheville 75

Charlotte 77

Fayetteville 79

Greensboro 77

New Bern 81

(continued )

Appendices 255

Table A.1. Continued

State City WB Temp (8F)

Raleigh 78

North Dakota Bismarck 72

Grand Rapids 75

Ohio Akron 75

Cincinnati 77

Columbus 77

Oklahoma Oklahoma City 78

Oregon Eugene/Portland 69

Pennsylvania Philadelphia 78

Pittsburgh 75

Rhode Island Providence 76

South Carolina Greenville 77

Columbia 78

Charleston 80

South Dakota Pierre 68

Rapid City 64

Sioux Falls 71

Tennessee Chattanooga 78

Knoxville 77

Memphis 78

Nashville 78

Texas Abilene 75

Amarillo 71

Austin 78

Dallas 78

El Paso 70

Houston 80

Laredo 79

San Antonio 79

Utah Salt Lake City 66

Vermont Burlington 74

Virginia Norfolk 80

Richmond 79

Roanoke 75

Washington District of Columbia 79

Washington Seattle 66

Yakima 67

West Virginia Charleston 76

Wisconsin Madison 76

Milwaukee 76

Wyoming Cheyenne 62

aFrom Table 1b, Chapter 27, 2001 ASHRAE Handbook, Fundamentals.

Appendices256

APPENDIX B: DRAFT SPECIFICATIONSAPPENDIX B1

POSITIVE DISPLACEMENT COMPRESSOR WATERCHILLERS

Part 1 GeneralQuality Assurance

ARI Compliance: Test and rate reciprocating chillers in accordance with ARI

Std. 550/590-98, “WaterChilling Packages Using the Vapor Compression

Cycle.”

ASHRAE Compliance:

Fabricate and install reciprocating compressor water chillers to comply

with ASHRAE 15 “Safety Code for Mechanical Refrigeration.”

Energy performance of equipment shall meet or surpass the requirements

of ASHRAE/IESNA 90.1-1999.

NEC Compliance: Comply with applicable NEC requirements pertaining to

electrical power and control wiring for construction and installation of

reciprocating chillers.

ANSI Compliance: Comply with ANSI B9.1 safety code requirements

pertaining to unit construction of reciprocating chillers.

ASME Compliance: Construct and test reciprocating aircooled liquid

chiller in accordance with ASME Boiler and Pressure Vessel Code,

Section 8.

NEMA Compliance: Provide high-efficiency motors for reciprocating

chillers that comply with NEMA Stds. Pub/No.’s MG 1, 2, 3, 10,

and 11.

UL Compliance: Comply with applicable requirements of UL 465, “Central

Cooling Air Conditioners,” pertaining to construction and installation of

reciprocating chillers.

Submittals

General: Submittals shall demonstrate compliance with technical requirements

by reference to each subsection of this specification. Where a submitted item does

not comply fully with each and every requirement of the Specifications, the

submittal shall clearly indicate such deviations. Identification requirements for

non-complying features of items are very specific.

Appendices 257

Manufacturer’s Data: Submit manufacturer’s technical product data, including

rated capacities for chillers indicated, weights (shipping, installed, and

operating), furnished specialties and accessories, and rigging, installation, and

start-up instructions.

Part 2 ProductsWater-Cooled Chillers

General: Provide factory-assembled and tested multiple compressor water-

cooled liquid chillers as indicated, consisting of two or more compressors,

condenser, evaporator, thermal expansion valve, and control panel. Provide

capacity and electrical characteristics as scheduled.

Refrigerant: Provide full operating charge of refrigerant (R-22, R-407c, or

R-410a) and oil.

Evaporator: Provide shell-and-tube design with seamless copper tubes roller

expanded into tube sheets. Design, test, and stamp for refrigerant side working

pressure of 225 PSIG minimum, and water side working pressure of 150 PSIG

minimum, in accordance with ASME Pressure Vessel Code. Insulate with 1/200minimum flexible unicellular insulation with maximum K-value of 0.28. Provide

water drain connection and bulb wells for temperature controller and low-

temperature cutout.

Multiple-Compressor Units: Provide independent multiple refrigerant circuits

with gasketed evaporator heads.

Condenser: Provide shell-and-tube design with seamless integral fin copper

tubes expanded into tube sheets, with tubes mechanically cleanable and

replaceable through removable headers. Design, test, and stamp for refrigerant

side working pressure of 300 PSIG minimum, and water side working pressure

of 150 PSIG minimum, in accordance with ASME Pressure Vessel Code.

Provide safety relief valve with pressure rating not to exceed condenser shell

side working pressure. Provide integral subcooler circuit.

Multiple-Compressor Units: Provide independent multiple refrigerant

circuits.

Compressors: Provide direct-drive, 1750RPM, accessible hermetic, multi-

cylinder reciprocating compressor with multistep capacity control. Provide

crackcase heater and suction strainer. Mount compressor on vibration isolators

within chiller housing.

Lubrication: Provide reversible, positive-displacement oil pump. Provide oil

charging valve, oil level sight glass, oil filter, andmagnetic plugor strainer.

Motor: Provide suction gas-cooled motor with high temperature protection.

Appendices258

Capacity Modulation: Provide step capacity control by means of cylinder

unloaders and/or compressor staging, from return water temperature.

Refrigerant Circuit: Provide the following for each refrigerant circuit:

Liquid line solenoid valveFilter dryer

Liquid line sight glass and moisture indicator

Thermal expansion valve

Compressor discharge service valve

1/400 flare charging port

Insulated suction line

Control Panel: Provide control panel, factory-wired for only single-point power

connection.

Provide the Following for Each Compressor:

High-pressure cutout

Motor protection

Provide the Following for Each Refrigerant Circuit:

Low-pressure cutout

Oil pressure cutout

Provide the Following for Each Unit:

Temperature controller

Low-temperature cutout

Sequence device (for multiple compressor units)

Terminal strip

Reset relay

Provide the Following for Each Compressor:

Power controls for part-wind or across-the-line start

Control power transformer or terminal block for 115 v control voltage

Compressor starter relay

Nonrecycling compressor overload device

Accessories: Provide the following accessories:

Vaporproof chilled water flow switch

Suction, discharge, and oil pressure gauges for each compressor

Compressor suction and discharge service valves

Spring vibration isolators

Antirecycle timer if protection against liquid slugging is not provided

Manual pumpdown switch

Appendices 259

Outdoor Air-Cooled Chillers

General: Provide factory-assembled and tested multi-compressor outdoor air-

cooled liquid chillers as indicated, consisting of two or more compressors,

evaporator, condensers, thermal expansion valves, and control panels. Provide


Refrigerant: Provide full operating charge of refrigerant (R-22 orR-410a) and oil.

Housing: Provide manufacturer’s standard equipment housing construction,

corrosion protection coating, and exterior finish. Provide removable panels

and/or access doors for inspection and access to internal parts and components.




minimum, in accordance with ASME Pressure Vessel Code. Provide one water

pass with series of internal baffles. Insulate with 1/200 minimum flexible

unicellular insulation with maximum K-value of 0.28. Provide water drain

connection and bulb wells for temperature controller and low-temperature cutout.

Heater Tapes: Provide electrical resistance heater tape on evaporator to

protect against freezing at 2208F (298C) ambient at no-flow condition.

Multiple-Compressor Units: Provide independent multiple refrigerant circuits

with gasketed evaporator heads.

Condenser: Construct coils with configured aluminum fins mechanically bonded

to seamless copper tubing. Provide integral subcooling circuit with liquid

accumulators. Leak test coils with air under water at 425 PSIG air pressure.

Provide protective grilles over exposed coil faces.

Multiple-Compressor Units: Provide multiple circuited condenser coils.

Condenser Fans: Provide propeller fans, direct or belt driven, draw-

through design, statically and dynamically balanced. Provide protective

grille over air discharge.

Low-Ambient Control: Provide head pressure control, designed to operate

at temperatures down to 08F (188C).

Compressors: Provide direct drive 1750RPM, multicylinder reciprocating

compressors with crankcase heater, either semihermetic or hermetic, but with

minimum steps of capacity control as scheduled provided by cylinder unloading

or compressor staging or combination of both. Mount compressors on vibration

isolators within chiller housing.

Lubrication: Provide oil pump, oil filter, oil level sight glass, and oil

charging valve.

Appendices260

Capacity Modulation: Provide step-control by means of cylinder unloading

and/or compressor staging, from return water temperature.

Refrigerant Circuit: Provide for each refrigerant circuit the following:

Liquid line solenoid valve

Filter dryer

Liquid line sight glass



Suction and discharge valves

Controls Panels: Provide weathertight control panels, factory-wired for external

connection only. Provide the following controls:

Power controls for starter

Control power transformer for 115V control voltage

Terminal strip

Pumpdown control relay


Reset relay

Nonrecycling compressor overload relay


Low-pressure cutout

Oil pressure cutout, except for low oil pressure compressor systems


Chilled water temperature controller



Suction and discharge gages, for each compressor

Oil pressure gages except for hermetic compressors


Part 3 ExecutionInstallation of Chillers

General: Install chillers in accordance with manufacturer’s written instructions.

Install units plumb and level, firmly anchored in locations indicated; maintain

manufacturer’s recommended clearances.

Appendices 261

Relief Piping: Provide relief piping from indoor chiller’s refrigerant pressure

relief rupture disc on chiller to atmosphere; size piping as recommended by

chiller manufacturer, and terminate with gooseneck facing down.

Start-up chillers, in accordance with manufacturer’s start-up instructions, and in

presence of manufacturer’s technical representative. Test controls and

demonstrate compliance with requirements. Replace damaged, or malfunction-

ing, controls and equipment and retest.

Do not place chillers in sustained operation prior to initial balancing of

mechanical systems that interface with the reciprocating chillers.

Provide services of manufacturer’s technical representative to instruct Owner’s

personnel in operation and maintenance of reciprocating chillers. Review with the

Owner’s personnel the data contained in the Operating andMaintenanceManuals.

APPENDIX B2

ROTARY WATER CHILLERSPart 1 General

Quality Assurance

ARI Compliance (up to 1600 tons): Fabricate, test, and rate chillers to comply

with ARI Standard 550/590-98 “Water-Chilling Packages Using the Vapor

Compression Cycle”.

ASHRAE Compliance:

Fabricate and install rotary water chillers to comply with ASHRAE 15

“Safety Code for Mechanical Refrigeration.”

Energy performance of equipment shall meet or surpass the requirements

of ASHRAE/IESNA 90.1-1999.

UL Compliance: Fabricate rotary water chillers to comply with UL 465

“Central Cooling Air Conditioners.”

ASME Compliance: Fabricate and stamp rotary water chillers to comply

with ASME Boiler and Pressure Vessel Code, Section VIII, Division 1.

Submittals




submittal shall clearly indicate such deviations. Identification requirements for

non-complying features of items are very specific.

Appendices262

Manufacturer’s Data: Submit manufacturer’s product data, including rated

capacities, weights (shipping, installed, and operating), furnished specialties and

accessories, and installation and start-up instructions.

Part 2 ProductsWater-Cooled Chillers

Description: Packaged, factory-assembled, hermetic or open-drive type high or

medium pressure water chillers consisting of centrifugal or rotary screw

compressor(s), compressor motor(s), motor starter(s), evaporator, condenser,

controls and panels including gages and indicating lights, auxiliary components,

and accessories.

Centrifugal Compressor(s):

Provide single or dual single-stage centrifugal compressors with capacity,

peak power input, and ILPC as scheduled.

Shaft and Impeller Assembly: Carbon or forged steel shaft with cast high

strength aluminum alloy impellers, designed and assembled for no

critical speeds within operating range; and statically and dynamically

balanced.

Casing: Fine grain cast iron with gasket sealed casing joints.

Drive Assembly: Gear transmission integral with compressor and lubricated

through compressor lubrication system.

Lubrication System: Forced circulation type, with positive displacement

submerged pump and replaceable oil filter; complete with an automatic

oil heater designed to separate refrigerant from oil, and oil cooler if

required for proper performance. System shall provide positive pressure

lubrication of journals, bearings and seals (if any), during start-up,

operation, and coast-down of chiller, including power interruptions. On

two compressor units provide redundant oil pump.

Capacity Control: Provide variable guide vanes to provide stable operation

without surge, cavitation, or vibration from 100 to 15% of full load

capacity, without hot-gas bypass.

Rotary Screw Compressor(s):

General: Compressor shall be rotary twin screw type operating at not more

than 3750 rpm.

Construction: Casing shall be fine grain cast iron with gasket-sealed casing

joints, machined to provide minimum, uniform clearance for the rotors.

Rotors shall be forged steel with asymmetric profiles. Provide bearings

Appendices 263

to adequately contain both radial and axial loading and maintain accurate

rotor positioning at all pressure ratios. Provide check valve on

compressor discharge to prevent rotor backspin.

Open-Drive Shaft Seal: Spring loaded, precision carbon ring, high

temperature elastomer “O” ring static seal, and stress-relieved, precision

lapped collars. The entire shaft seal cavity shall be at low pressure, and

shall be vented to the oil drain from the compressor.

Lubrication System: Provide oil separator on discharge of compressor to

provide minimum 99% separation prior to refrigerant entering

condenser. Main oil reservoir shall be located so as to provide gravity

lubrication of rotor and rotor bearings during start-up and coast-down;

system differential pressure shall provide proper oil flow during normal

operation.

Capacity Control: Achieve by use of a slide valve, which shall provide

fully modulating capacity control from 100% to 10% of full load. The

slide valve shall be actuated by oil pressure, controlled by external

solenoid valves via the chiller integral controls.

Compressor Motor(s):

Motor and Accessories (Hermetic Type): Hermetically sealed, continuous

duty, single speed, squirrel cage, induction type; full load operation of the

motor shall not exceed nameplate rating; rotor shaft shall be heat treated

carbon steel and designed such that the first critical speed iswell above the

operating speed. Provide for removal of the stator for service or

replacement without breaking the main refrigerant piping connections.

Motor and Accessories (Open-Drive Type): Two-pole, continuous duty,

squirrel cage induction type, and shall have an open drip-proof

enclosure. Motor full-load amperes at design conditions shall not exceed

motor nameplate (FLA). Motor shall be designed for use with the type

starter specified. Motor shall be factory mounted and directly supported

by the compressor in providing controlled alignment.

Compressor Motor Starter(s): Provide unit-mounted or free-standing factory-

wired starters. All power and control conductors between starters and other

chiller components, whether unit mounted or free standing, shall be sized,

furnished, and installed, complete and ready for operation. Starter shall

incorporate a single point for chiller power connection. All chiller heaters,

pumps, control circuits, and other miscellaneous items and accessories shall be

supplied directly from the starter without dependence on external power from

other sources. Where ancillary items require a different supply voltage from the

chiller motor, a suitable transformer(s) and appropriate wiring shall be provided

Appendices264

as a part of the chiller starter to supply these items. Starter shall be reduced

voltage, either closed transition wye-delta or solid-state type. Starter shall be non-

reversing, enclosed in a suitable NEMA-1 enclosure, and shall have as a

minimum the following features:

Self-contained disconnecting means to disconnect all power sources from

the chiller and related accessories.

Electronic internal self-contained monitor system to protect against

3-phase and single phase overloads during running and starting, loss of

phase, phase reversal, phase unbalance voltage or current, overvoltage,

and undervoltage.

A total run-time meter mounted on the starter door.

An electronic metering module mounted on the starter door providing the

following instantaneous, demand, and energy information:

Instantaneous: Digital readings of line and phase voltages, line and

phase currents, total Watts, total VARS, total kVA,

power factor, and line frequency.

Demand: Digital readings of line and phase currents, and total Watts.

Energy: Digital readings of accumulated Watthours and accumulated

VAR-hours.

Evaporator and Condenser:

Shell and Water Boxes: Fabricated from welded carbon steel plate. Provide

150 psig maximum working pressure water boxes and nozzle

connections. Provide vents, drains, and covers in water boxes to permit

tube cleaning within the space shown on the Drawings. Provide suitable

tappings in the water boxes and nozzles for control sensors, gages, and

thermometers.

Water Heads: Fabricated steel water heads with integral water connections.

Tube Sheets: Fabricated of carbon steel sheets welded to the shell and

drilled for tubes. Include intermediate tube support sheets as required to

prevent tube vibration.

Tubes: Individually replaceable, finned, seamless copper tubes; removable

from either end of the heat exchanger without effecting strength and

durability of the tube sheets and without causing leakage in adjacent

tubes. Expand ends of tubes in tube sheets and intermediate tube support

sheets for tight fit to prevent vibration of tubes. Provide suitable baffles

or distributing plates in condenser tubes to evenly distribute refrigerant

discharge gas on heat transfer tubes.

Appendices 265

Pressure Limiting and Pressure Relief Devices: Manufacturer’s standard

complying with ASHRAE 15. Provide refrigerant charging and transfer

connections, pressure relief device on the evaporator to prevent

excessive pressure in refrigerant side, and means to sense refrigerant

pressure or temperature. Pressure relief valve shall close once pressure is

reduced below setpoint pressure.

Accessories:

Pump-out System: Include compressor and drive, piping, wiring, motor

starter, and manual isolation valves so that entire refrigerant charge can

be contained and isolated in either condenser or evaporator, as required.

Where multiple chillers of the same type and utilizing the same refrigerant

are provided, a single portable pump-out unit may be provided in lieu of

individual units with each chiller.

Refrigerants: Package shall utilize R-22, R-134a, or R-410a.

Controls and Safeties:

Refrigerant Flow Control Devices: Provide refrigerant flow control devices

between evaporator and condensers (and elsewhere as required) to

regulate refrigerant flow at volume and pressure required to maintain

evaporator liquid refrigerant at level sufficient to keep cooler heat

transfer tubes adequately wetted through full range of chiller operation.

Design devices to permit chiller operation at scheduled conditions, and to

allow condenser entering water temperature to decrease to minimum

permissible temperature or 18F above return chilled water temperature.

Safety Controls: Design cutouts to operate independently and factorywire to

control panel. Design controls to stop compressor motor in event of low

refrigerant pressure or temperature in evaporator, high condenser

pressure, high compressor discharge temperature, low evaporator leaving

water temperature (freezestat), high motor temperature, high bearing

temperature, low oil pressure, high oil temperature, compressor motor

overcurrent or over voltage, or power interruption. Design each cutout to

require manual restarting of compressor in event of safety failure but

allowing automatic restart after power interruption. Provide anti-recycle

timer for limiting compressor motor restarts at scheduled time intervals.

Provide normally closed dry contact to indicate “chiller failure,”

connected to building control system, that opens in event of any safety

control activation.

Operational Controls: Provide controls to ensure that compressor will start

only under unloaded condition.

Appendices266

Provide sequencing controls to ensure lubrication of compressor motor

bearings and seals (if any). Sequence as follows:1. Run lubrication system oil pump so that compressor motor bearing

is lubricated before start-up.

2. Start compressor motor.

3. Provide lubrication during coast-down after compressor motor

shutdown.

Design controls to automatically restart compressor after power failure

interruptions, providedminimum timebetween starts has been compliedwith.

Control Panel: Factory-mounted and wired. Provide gages or meters to

indicate low refrigerant pressure in evaporator, high condenser pressure, and

low oil pressure.Provide switches and pushbuttons designed to permit indicated

operations including the following:Manual and automatic operation of oil pump.

Manual and automatic operation of oil separator heater.

Provide pilot lights or visual flag switches for indicated operations and

cutouts including the following:

Oil pump operation

Low chilled water temperature cutout

Low evaporator refrigerant pressure or temperature cutout

High condenser pressure cutout

High motor winding temperature cutout

Low oil pressure cutout

Motor overload cutout

Diagnostics: Provide a diagnostic module capable of indicating all

lockout conditions specified above, plus recording the elapsed time

(prealarm to alarm), the operating conditions of the compressor

motor (amperes), refrigerant temperatures and pressures, and chilled

and condenser water temperatures (entering and leaving) at the time

of lockout.

DDC Interface: Where a direct digital control system is provided, the

chiller control panel shall include necessary additional input/output

devices to allow interface to the DDC system, as follows:4-20ma AI for CHWS temperature monitoring

4-20ma AI for CHWR temperature monitoring

4-20ma AI for CDWS temperature monitoring

4-20ma AI for CDWR temperature monitoring

4-20ma AO for CHWS temperature control (setpoint)

Appendices 267

4-20ma AO for chiller loading control. Provide adjustable

control device to decrease compressor motor power when

power demand reaches selected percentage of full load

power demand, adjustable from 40 to 100% continuously.

Design mechanism to control operation by monitoring motor

amperage and modulating inlet vanes. Provide interface to

accept exterior 4-20mA control signal for set-point

adjustment from building control system.

Contact closure DI for chiller alarm

Insulation: Insulate evaporators and all other cold surfaces to prevent

condensation, with ambient humidity of 75 percent and drybulb temperature of

90 deg. F, no air movement.

Outdoor Air-Cooled Chillers

General: Provide factory-assembled and tested multi-compressor outdoor air-

cooled liquid chillers as indicated, consisting of at least two compressors,

evaporator, condensers, thermal expansion valves, and control panels. Provide


Refrigerant: Provide full operating charge of R-22, R-410a, or R-134A

refrigerant and oil.

Housing: Provide manufacturer’s standard equipment housing construction,

corrosion protection coating, and exterior finish. Provide removable panels

and/or access doors for inspection and access to internal parts and components.




minimum, in accordance with ASME Pressure Vessel Code. Provide one water

pass with series of internal baffles. Insulate with 1/200 minimum flexible

unicellular insulation with maximum K-value of 0.28. Provide water drain

connection and bulb wells for temperature controller and low-temperature cutout.

Heater Tapes: Provide electrical resistance heater tape on evaporator to

protect against freezing at 208F (298C) ambient at no-flow condition.

Multiple-Compressor Units: Provide independent multiple refrigerant

circuits with gasketed evaporator heads.

Condenser: Construct coils with configured aluminum fins mechanically bonded

to seamless copper tubing. Provide integral subcooling circuit with liquid

accumulators. Leak test coils with air under water at 425 PSIG air pressure.

Provide protective grilles over exposed coil faces.

Appendices268

Multiple-Compressor Units: Provide multiple circuited condenser coils.

Condenser Fans: Provide propeller fans, direct or belt driven, draw-

through design, statically and dynamically balanced. Provide protective

grille over air discharge.

Low-Ambient Control: Provide head pressure control, designed to operate

at temperatures down to 08F (188C).

Rotary Screw Compressor(s):

General: Compressor shall be rotary twin screw type operating at not more

than 3750 rpm.

Construction: Casing shall be fine grain cast iron with gasket-sealed casing

joints, machined to provide minimum, uniform clearance for the rotors.

Rotors shall be forged steel with asymmetric profiles. Provide bearings

to adequately contain both radial and axial loadings and maintain

accurate rotor positioning at all pressure ratios. Provide check valve on

compressor discharge to prevent rotor backspin.

Open-Drive Shaft Seal: Spring loaded, precision carbon ring, high

temperature elastomer “O” ring static seal, and stress-relieved, precision

lapped collars. The entire shaft seal cavity shall be at low pressure, and

shall be vented to the oil drain from the compressor.

Lubrication System: Provide oil separator on discharge of compressor to

provide minimum 99% separation prior to refrigerant entering condenser.

Main oil reservoir shall be located so as to provide gravity lubrication of

rotor and rotor bearings during start-up and coast-down; systemdifferential

pressure shall provide proper oil flow during normal operation.

Capacity Control: Achieve by use of a slide valve, which shall provide

fully modulating capacity control from 100% to 10% of full load. The

slide valve shall be actuated by oil pressure, controlled by external

solenoid valves via the chiller integral controls.

Refrigerant Circuit: Provide for each refrigerant circuit the following:

Liquid line solenoid valve

Filter dryer

Liquid line sight glass



Suction and discharge valves

Appendices 269

Controls Panels: Provide weathertight control panels, factory-wired for external

connection only. Provide the following controls:

Across-the-line starters

Power controls for starter

Control power transformer for 115V control voltage

Terminal strip

Pumpdown control relay


Reset relay

Nonrecycling compressor overload relay


Low-pressure cutout

Oil pressure cutout, except for low oil pressure compressor systems


Chilled water temperature controller



Suction and discharge gages, for each compressor

Oil pressure gages except for hermetic compressors


Part 3 ExecutionInstallation of Chillers

General: Install chillers in accordance with manufacturer’s written instructions.

Install units plumb and level, firmly anchored in locations indicated; maintain

manufacturer’s recommended clearances.

Relief Piping: Provide relief piping from indoor water-cooled chiller refrigerant

pressure relief rupture disc to atmosphere; size piping as recommended by chiller

manufacturer, and terminate with gooseneck facing down.

Include Factory Performance Test Only when Required

Factory Performance Test

General: Manufacturer shall conduct witnessed performance acceptance test for

each typical machine to verify the performance submitted. The test will be in

accordance with ARI 550/590-98with the following exceptions:

Capacity will have no tolerances. Capacity must meet or better required

tonnage.

Heat balance must be verified at 3% or below during test conditions.

Appendices270

Test Site: Test shall be conducted at an approved ARI Certified Test Facility of

Manufacturer’s factory. Instrumentation used for testing must be calibrated within

6months of the test date and traceable to the National Bureau of Standards. All

documentation verifying NBS traceability shall be included in a bound folder for

presentation to the Owner’s representative. The Owner’s representative may elect

to contact ARI for verification of performance and test conditions.

Test Conditions for Fouling Factor Adjustment: Chiller shall be tested with

water temperature and adjustment cooler/condenser per standard ARI 550-90 to

simulate specified fouling vs. no fouling during test. Verification of this

procedure will require inside surface area and number of tubes per vessel. This

information is to be submitted prior to the test for formula verification of fouling

per ARI 550-90 if so requested by Engineer.

Test Points: Performance test shall be a four point test, 25, 50, 75, and 100%

load.

Performance Penalty Clause: The chiller will be accepted if the test indicates

that all conditions of the performance test meet the specified conditions. If the

performance test results are outside of specified tolerances, the manufacturer will

be allowed to make repairs and retest; and the manufacturer will assume all

expenses incurred for the Owner’s representatives to witness the retest.

If after the second test the machines do not meet the test requirements, then

an energy cost penalty, of $4000 per kW above test requirements, will be

deducted from the manufacturer’s contract. In the case of an ARI Certified

Performance chiller, ARI will be notified of any test failures. If there is more than

one typical machine, the energy penalty will be the penalty of the tested machine

times the number of typical machines on the project. The total energy penalty will

be sum of all machines on the project whether tested or typical.

Demonstration

Start-up chillers, in accordance with manufacturer’s start-up instructions, and

provide the service of a factory authorized service representative to provide start-

up service and to demonstrate and train the Owner’s maintenance personnel as

specified below. Evacuate, dehydrate, and charge with specified refrigerant, and

leak test in accordance with manufacturer’s instructions. Perform lubrication

service, including filling of reservoirs, and confirming that lubricant is of quantity

and type recommended by manufacturer. Test controls and demonstrate

compliance with requirements. Replace damaged, or malfunctioning, controls

and equipment and retest.

Do not place chillers in sustained operation prior to initial balancing of

mechanical systems which interface with the reciprocating chillers.

Appendices 271

Training: Train the Owner’s maintenance personnel on start-up and shut down

procedures, troubleshooting procedures, and servicing and preventive mainten-

ance schedules and procedures. Review with the Owner’s personnel, the data

contained in the Operating and Maintenance Manuals.

APPENDIX B3

INDUCED DRAFT COOLING TOWERS, CROSSFLOW ORCOUNTERFLOW, FACTORY-ASSEMBLED

Part 1—GeneralQuality Assurance

Provide Cooling Technology Institute (CTI) certification of tower cooling

capacity, based on performance tests, in accordance with CTI Standard 201.

Manufacturer Guarantee: Provide written manufacturer guarantee that the

tower will perform in complete compliance with performance requirements,

provided the tower is installed in accordance with the manufacturer’s

written instructions. This guarantee shall include the manufacturer providing

in-situ testing of the tower performance in accordance with CTI Standard ATC-

105, by a testing agency licensed by CTI, if required by the designer or owner.

Codes and Standards:

Testing Laboratory and NEMA Compliance: Provide Listed and Labeled

electric motors and electrical components required as part of factory-

fabricated cooling towers, which comply with NEMA Standards.

NFPA Compliance: Install cooling towers in accordance with NFPA 70

“National Electrical Code”.

Submittals




submittal shall clearly indicate such deviations.

Manufacturer’s Data:

Submit manufacturer’s technical product data, including rated capacities,

pressure drop, fan performance data, weights (shipping, installed, and

operating), and installation and start-up instructions.

Submit CTI performance certification.

Appendices272

Part 2—Products

General: Cooling tower shall be a complete factory-assembled, totally

noncombustible unit.

Structure: Structure shall be steel with galvanized finish consisting of minimum

2.35 oz per square foot of surface area. Assemble structural system, including

basin and casings, by providing bolted connections with fasteners having equal or

better corrosion-resistance than the materials fastened. Seal joints to make

watertight enclosure.

Certify tower structure for wind loading caused by 100mph wind from any

direction.

Provide rigging supports on structure for final rigging.

Casing: Provide fiberglass-reinforced polyester or minimum 22 ga. steel panels

with galvanized finish of minimum 2.35 oz of zinc per square foot of surface area

and fabricated to make tower watertight.

Basin:

Basin shall be constructed of minimum 20 ga. stainless steel, designed and

fabricated to contain and support the water and ensure water tightness:

Provide integral type collecting basin and sump with side outlet and with

connections for drain, overflow and water make-up. Drain connection

and basin operating level shall be designed to prevent vortexing and air

entrainment.

Wetted-Surface (Film) Fill: Provide vertical sheets of PVC or CPVC plastic

having flame spread rating of 5 in accordance with ASTM E84 and fabricated into

wave-formed configurations installed by manufacturer to assure maximum

wetted heat transfer surface.

Drift Eliminators: Provide PVC or CPVC plastic having flame spread rating of 5

in accordance with ASTM E84, which is fabricated into configuration to limit

drift loss to a maximum of 0.1% of the condenser water flow-rate. Drift

eliminators may be integral with the fill.

Louvers: Provide fiberglass reinforced plastic or polyvinyl chloride plastic of

sufficient thickness and rigidity to prevent visible sagging or fluttering.

Wet Deck Water Distribution:

Cross-flow Tower: Provide stainless steel open basin with gravity-flow

plastic metering orifices providing even distribution of water over fill.

Provide flow control valves for balancing flow to each distribution basin

where multiple inlets are required. Provide removable fiberglass-

Appendices 273

reinforced polyester panels to prevent debris from entering the wet deck

and to inhibit algae growth by eliminating sunlight.

Counter-flow Tower: Provide the to ensure even distribution of water over

wetted-surface-fill.Schedule 40 PVC pipe header and removable schedule 40 PVC pipe

branches.

Nozzles: Provide removable plastic, brass, or ceramic nozzles with a

maximum 12 psig inlet pressure required.

Inlet Screens: Provide stainless steel mesh, mounted in removable frames by

manufacturer.

Select One of the Following Types of Basin Heaters

Basin Heaters: Provide steam injection heaters, with 2-position steam control

valve and thermostat set at 408F.

Provide basin heaters sized by manufacturer to maintain basin water at

408F at ambient temperature of 08F and wind velocity of 15mph.

Basin Heater: Equivalent to Penberthy Model NWK, minimum 20 psig

steam pressure rating.

Basin Heaters: Provide electric immersion heaters including thermostat, low-

water cutout contactor, and transformers, in weatherproof enclosure.



Basin Heater: Screw-plug type, 2-1/2 inch pipe thread, electric immersion

heater. Heaters specified for 3-phase service shall be provided with

necessary contactors. Heater element shall be copper sheathed, brazed to

brass screw plugs. Multiple heaters shall be provided, wired in parallel,

as required to meet indicated capacity. Heater shall be equivalent to

Chromolox Type ARMT.

Select One of the Following Methods of Basin Water Level Control

Float ValveWater Level Control: Provide float-activated make-up water valve.

Float shall be fully adjustable as to depth control and angle to provide tight

shutoff. Set-screw adjustments are not acceptable.

Electronic Water Level Control: Provide electric water level control package

including conductance-actuated level controller and slow closing solenoid make-

up valve. Level control shall provide for independently adjustable high operating

level to close make-up water valve and low operating level to open make-up

Appendices274

water valve. Additional alarm contacts for high water and low water levels shall

be provided.

Fan and Drive: Provide propeller fan with cast-aluminum adjustable pitch blades

driven by one of the following methods. Fan shall be statically and dynamically

balanced and hub shall be keyed to ground and polished steel shaft:

Gear Drive: Provide right angle, industrial duty, oil-lubricated geared

speed reducer with shafts to connect motor and fan. Gear drive shall be

selected in accordance with American Gear Manufacturers Association

Standard 420 for continuous duty with minimum 1.5 service factor. Gear

bearings shall be heavy duty grease lubricated ball or roller bearings with

a “Basic Rating Life”, L10, as defined by the American Bearing

Manufacturers Association, of at least 100,000 hours at maximum

operating speed and horsepower

Belt drive: Provide cast aluminum fan and motor sheaves with 1-piece

multi-groove neoprene/polyester drive belt.

Fan Bearings: Bearings shall be “air handling quality” heavy duty grease

lubricated ball or roller bearings with a “Basic Rating Life”, L10, as defined by

the American Bearing Manufacturers Association, of at least 100,000 hours at

maximum operating speed and horsepower.

Motor: Provide two-speed, one-winding, motor rated at 1800/900 rpm. Motor

located in the tower air stream may be TEAO type, while motor located out

of the tower air steam shall be TEFC type. ODP motors are unacceptable.

Vibration Cutout Switch: Provide switch to de-energize fan motors if excessive

vibration occurs due to fan imbalance.

Service Access:

Cross-flow Tower: Provide minimum 2400 wide by 3000 high access doors onboth end walls to provide access to basin, fan plenum, motor, drive, and

fill. Provide minimum 1800 wide galvanized steel grated walkway across

the tower above the basin overflow level. If the tower is installed so that

the bottom of the access door is more than 500– 000 above the roof or

grade, provide an access door platform on each side with ladder rigidly

attached to the tower.

Top of Tower Handrail and Ladder: Where scheduled on the drawings,

provide top of tower galvanized steel structural tubing handrail,

complete with knee rail and toeboard, constructed in accordance with

OSHA guidelines. Provide ladder from roof or grade to top of tower.

Provide ladder safety cage if the top of tower is 20’– 0” of greater above

grade or roof.

Appendices 275

Apply phosphatized pretreatment on any zinc-coated surfaces that have not been

mill-phosphatized or polymer-coated. Apply gasoline-soluble rust preventive

compound on ferrous parts that cannot be galvanized, including shafts and

machined parts.

Finish components with zinc-coated metal surfaces by applying to metal

surfaces not galvanized, a zinc rich paint that has been tested and accepted by a

testing laboratory as being equivalent to galvanized steel.

Part 3—ExecutionInstallation

General: Install cooling towers where indicated, in accordance with equipment

manufacturer’s written instructions and with recognized industry practices, to

ensure that cooling towers comply with requirements and serve intended

purposes.

Support: Install cooling tower on structural supports as indicated on the

drawings. Anchor cooling tower to stand with removable fasteners. Tower shall

be installed level. Where multiple tower cells are indicated, the entire tower

assembly shall be leveled end-to-end within 0.2500.Electrical: Provide fan motor disconnect switch in NEMA 3R enclosure rigidly

attached to the cooling tower and located immediately adjacent to the primary

access door. Provide service clearance required by NFPA 70.

Start-Up

Condenser Water System Start-Up: Comply with all manufacturer instructions

forfillingandstart-upof thecondenserwater system,butnot less than the following:

Condenser Water Pump

1. Check pump installation, including mountings, vibration isolators

and connectors, and piping specialties (valves, strainer, pressure

gauges, thermometers, etc.)




5. Turn shaft by hand to make sure the pump and motor turn

freely.


Cooling Tower1. Clean all tower surfaces, flush and clean the wet deck and basin.

Appendices276

2. Clean the basin strainer



5. Test and adjust belt drive (if installed):Adjust belt(s) tension.

Check and adjust belt(s) alignment.

6. Test and adjust gear drive (if installed):Fill oil reservoir.

Check shaft alignment

Check couplings for bolt tightness, excess play, etc.



9. Run fan(s) for a short period and check for unusual noises and/or

vibration. Verify that motor amps are in accordance with

manufacturer’s data.


chemically cleaned.

11. Fill the basin and piping to the manufacturer’s recommended

basin operating level.

12. Run condenser water pump for a short period:Verify that pump motor amps are within motor nameplate rating.

Test the wet deck distribution, fill, and basin for proper water flow.

Check basin for vortexing.


14. For multi-cell installations:Ensure that automatic isolation valves function properly.

Balance condenser water flow to and from each cell.

15. Balance equalizer line to maintain proper basin operating level in

each cell.


position.

Controls


2. Test operation of bypass control valve and adjust for 70-degree

rotation.

3. Check controller set-point.




7. Test operation of basin level controls, including high and low level

alarms.

Appendices 277


Water Treatment

1. Check that all pH and/or conductivity probes are properly

installed.



Operate the condenser water system to maintain set-point temperature under its

automatic controls for at least 24 hours. All aspects of condenser water pump,

cooling tower, controls, and water treatment shall be checked and evaluated

frequently during this period and any unusual noises or vibration, erratic

performance or operation, or other problems shall be repaired before being placed

into fulltime service.

Train the Owner’s maintenance personnel on start-up and shutdown


ance schedules and procedures. Review with the Owner’s personnel the data


APPENDIX B4

FORCED DRAFT COOLING TOWERS, CROSS-FLOW ORCOUNTER-FLOW, FACTORY-ASSEMBLED

Part 1—GeneralQuality Assurance

Provide Cooling Technology Institute (CTI) certification of tower cooling

capacity, based on performance tests, in accordance with CTI Standard 201.

Manufacturer Guarantee: Provide written manufacturer guarantee that the

tower will perform in complete compliance with performance requirements,

provided the tower is installed in accordance with the manufacturer’s written

instructions. This guarantee shall include the manufacturer providing in-situ

testing of the tower performance in accordance with CTI Standard ATC-105, by a

testing agency licensed by CTI, if required by the designer or owner.

Codes and Standards:

Testing Laboratory and NEMA Compliance: Provide Listed and

Labeled electric motors and electrical components required as

part of factory-fabricated cooling towers, which comply with

NEMA Standards.

Appendices278

NFPA Compliance: Install cooling towers in accordance with NFPA

70 “National Electrical Code”.

Submittals




submittal shall clearly indicate such deviations.

Manufacturer’s Data:

Submit manufacturer’s technical product data, including rated

capacities, pressure drop, fan performance data, weights

(shipping, installed, and operating), and installation and start-

up instructions.

Submit CTI performance certification.

Part 2—Products

General: Cooling tower shall be a complete factory-assembled, totally

noncombustible unit.

Structure: Structure shall be steel with galvanized finish consisting of minimum

2.35 oz per square foot of surface area or extruded PVC. Assemble structural

system, including basin and casings, by providing bolted connections with

fasteners having equal or better corrosion-resistance than the materials fastened.

Seal joints to make watertight enclosure.

Certify tower structure for wind loading caused by 100mph wind from any

direction.

Provide rigging supports on structure for final rigging.

Casing: Provide fiberglass-reinforced polyester or minimum 22 ga. steel panels

with galvanized finish of minimum 2.35 oz of zinc per square foot of surface area

and fabricated to make tower watertight.

Basin:

Cross-flow tower basin shall be constructed of minimum 20 ga. stainless

steel, designed and fabricated to contain and support the water and

ensure water tightness. Provide integral type collecting basin and sump

with side outlet and with connections for drain, overflow and water

make-up. Drain connection and basin operating level shall be designed

to prevent vortexing and air entrainment.

Appendices 279

Counter-flow tower basin shall be one of the following, as applicable:

Basin constructed of minimum 20 ga. stainless steel, designed and

fabricated to contain and support the water and ensure water

tightness. Provide integral type collecting basin and sump

with side outlet and with connections for drain, overflow

and water make-up. Drain connection and basin operating

level shall be designed to prevent vortexing and air

entrainment.

Water collection system and one or more exterior sumps designed to

collect and contain the tower cold water. These components

shall be constructed of stainless steel, fiberglass reinforced

plastic, or other non-corrosive material.

Wetted-Surface (Film) Fill: Provide vertical sheets of PVC or CPVC plastic

having flame spread rating of 5 in accordance with ASTM E84 and fabricated

into wave-formed configurations installed by manufacturer to assure maximum

wetted heat transfer surface.

Drift Eliminators: Provide PVC or CPVC plastic having flame spread rating of 5

in accordance with ASTM E84, which is fabricated into configuration to limit

drift loss to a maximum of 0.1% of the condenser water flow-rate. Drift

eliminators may be integral with the fill.

Wet Deck Water Distribution:

Cross-flow Tower: Provide stainless steel open basin with gravity-flow

plastic metering orifices providing even distribution of water over fill.

Provide flow control valves for balancing flow to each distribution basin

where multiple inlets are required. Provide removable fiberglass-

reinforced polyester panels to prevent debris from entering the wet deck

and to inhibit algae growth by eliminating sunlight.

Counter-flow Tower: Provide following the to ensure even distribution of

water over wetted-surface-fill.

Schedule 40 PVC pipe header and removable schedule 40 PVC pipe

branches.

Nozzles: Provide removable plastic, brass, or ceramic nozzles with a

maximum 12 psig inlet pressure required.

Inlet Screens: Provide stainless steel mesh, mounted in removable frames by

manufacturer.

Select One of the Following Types of Basin Heaters

Appendices280

Basin Heaters: Provide steam injection heaters with 2-position steam control

valve and thermostat set at 408F.



Basin Heater: Equivalent to Penberthy Model NWK, minimum 20 psig

steam pressure rating.

Basin Heaters: Provide electric immersion heaters including thermostat, low-

water cutout contactor, and transformers, in weatherproof enclosure.



Basin Heater: Screw-plug type, 2-1/2 inch pipe thread, electric immersion

heater. Heaters specified for 3-phase service shall be provided with

necessary contactors. Heater element shall be copper sheathed, brazed to

brass screw plugs. Multiple heaters shall be provided, wired in parallel,

as required to meet indicated capacity. Heater shall be equivalent to

Chromolox Type ARMT.

Select One of the Following Methods of Basin Water Level Control

Float Valve Water Level Control: Provide float-activated make-up water valve.

Float shall be fully adjustable as to depth control and angle to provide tight

shutoff. Set-screw adjustments are not acceptable.

Electronic Water Level Control: Provide electric water level control package

including conductance-actuated level controller and slow closing solenoid make-

up valve. Level control shall provide for independently adjustable high operating

level to close make-up water valve and low operating level to open make-up

water valve. Additional alarm contacts for high water and low water levels shall

be provided.

Fans and Drives: Provide propeller fans with cast-aluminum hub and aluminum

or fiberglass reinforced plastic blades driven by one of the following methods.

Fan shall be statically and dynamically balanced and hub shall be keyed to

ground and polished steel shaft:

Direct Drive: Provide keyed, flexible coupling to connect motor shaft to fan

shaft.

Belt drive: Provide cast aluminum fan and motor sheaves with 1-piece

multi-groove neoprene/polyester drive belt.

Fan Bearings: Bearings shall be “air handling quality” heavy duty grease

lubricated ball or roller bearings with a “Basic Rating Life”, L10, as defined by the

Appendices 281

American Bearing Manufacturers Association, of at least 100,000 hours at

maximum operating speed and horsepower.

Motors: Provide two-speed, one-winding, motor(s) rated at 1800/900 rpm. Motor

located in the tower air stream may be TEAO type, while motor located out of the

tower air steam shall be TEFC type. ODP motors are unacceptable.

Vibration Cutout Switch: Provide switch to de-energize fan motors if excessive

vibration occurs due to fan imbalance.

Service Access:

Cross-flow Tower: Provide minimum 2400 wide by 3000 high access doors onboth end walls to provide access to basin, fan plenum, motor, drive, and

fill. Provide minimum 1800 wide galvanized steel grated walkway across

the tower above the basin overflow level. If the tower is installed so that

the bottom of the access door is more than 50 –000 above the roof or grade,provide an access door platform on each side with ladder rigidly attached

to the tower.

Top of Tower Handrail and Ladder: Where scheduled on the drawings,

provide top of tower galvanized steel structural tubing handrail, complete

with knee rail and toeboard, constructed in accordance with OSHA

guidelines. Provide ladder from roofor grade to topof tower. Provide ladder

safety cage if the top of tower is 200 – 00 of greater above grade or roof.

Apply phosphatized pretreatment on any zinc-coated surfaces that have not been

mill-phosphatized or polymer-coated. Apply gasoline-soluble rust preventive

compound on ferrous parts that cannot be galvanized, including shafts and

machined parts.

Finish components with zinc-coated metal surfaces by applying to metal

surfaces not galvanized, a zinc rich paint that has been tested and accepted by a

testing laboratory as being equivalent to galvanized steel.

Part 3—ExecutionInstallation

General: Install cooling towers where indicated, in accordance with equipment

manufacturer’s written instructions and with recognized industry practices, to

ensure that cooling towers comply with requirements and serve intended

purposes.

Support: Install cooling tower on structural supports as indicated on the

drawings. Anchor cooling tower to stand with removable fasteners. Tower shall

Appendices282

be installed level. Where multiple tower cells are indicated, the entire tower

assembly shall be leveled end-to-end within 0.2500.

Electrical: Provide fan motor disconnect switch in NEMA 3R enclosure rigidly

attached to the cooling tower and located immediately adjacent to the primary

access door. Provide service clearance required by NFPA 70.

Start-Up

Condenser Water System Start-Up: Comply with all manufacturer instructions

for filling and start-up of the condenser water system, but not less than the

following:

Condenser Water Pump:

1. Check pump installation, including mountings, vibration isolators

and connectors, and piping specialties (valves, strainer, pressure

gauges, thermometers, etc.)






Cooling Tower

1. Clean all tower surfaces, flush and clean the wet deck and basin.

2. Clean the basin strainer



5. Test and adjust belt drive (if installed):Adjust belt(s) tension.

Check and adjust belt(s) alignment.

6. Test and adjust gear drive (if installed):

Fill oil reservoir.

Check shaft alignment

Check couplings for bolt tightness, excess play, etc.



9. Run fan(s) for a short period and check for unusual noises and/or

vibration. Verify that motor amps are in accordance with

manufacturer’s data.


chemically cleaned.

Appendices 283

11. Fill the basin and piping to the manufacturer’s recommended

basin operating level.

12. Run condenser water pump for a short period:Verify that pump motor amps are within motor nameplate

rating.

Test the wet deck distribution, fill, and basin for proper water

flow.

Check basin for vortexing.


14. For multi-cell installations:Ensure that automatic isolation valves function properly.

Balance condenser water flow to and from each cell.

Balance equalizer line to maintain proper basin operating level

in each cell.


position.

Controls


2. Test operation of bypass control valve and adjust for 70-degree

rotation.

3. Check controller setpoint.




7. Test operation of basin level controls, including high and low level

alarms.


Water Treatment

1. Check that all pH and/or conductivity probes are properly

installed.



Operate the condenser water system to maintain setpoint temperature under its

automatic controls for at least 24 hours. All aspects of condenser water pump,

cooling tower, controls, and water treatment shall be checked and evaluated

frequently during this period and any unusual noises or vibration, erratic

performance or operation, or other problems shall be repaired before being placed

into fulltime service.

Train the Owner’s maintenance personnel on start-up and shutdown


Appendices284

ance schedules and procedures. Review with the Owner’s personnel the data


APPENDIX C

REFERENCES AND RESOURCES

WATER CHILLERSWater Chiller General References

ASHRAE Handbook, Heating, Ventilating, and Air-Conditionings Systems and

Equipment. ASHRAE, Chap. 38, 2000.

ASHRAE Handbook, Refrigeration. ASHRAE, Chaps. 41, 43, 1998.

Centrifugal Chiller Fundamentals, Application Guide AG 31-002, McQuay Air-

Conditioning; McQuay International, 2000.

Refrigerants

Ozone-Depleting Substances: Position Paper, American Society of Heating, Refrigerating,

and Air-Conditioning Engineers. ASHRAE, 2001.

Chiller Performance Factors

White Paper on ARI Standard 550/590-98; American Refrigeration Institute, 1998.

System Part Load Value: A Case for Chiller System Optimization Carrier Synopsis, (3),

No. 3.

New ARI Rating Allows More Accurate Chiller-Energy Specification, HVAC&R

Engineering Update. York International Corporation, 1998.

Nowakowski, G.; Busby, R. Advances in Natural Gas Cooling. ASHRAE J. April:47,

2001.

Chiller Configuration and Control

Off-Design Chiller Performance. Engineers Newsletter 25(5), The Trane Company 1996.

Asymmetry as a Basis of Design. Engineers Newsletter 28(4), The Trane Company 1999.

Variable-Primary-Flow Systems. Engineers Newsletter 28(3), The Trane Company 1999.

Benson, J.W. Designing Chilled Water Systems for the 21st Century. The Trane Company

(Semiconductor FabTech, 9th Ed., 1998).

Coad, W.J. A Fundamental Perspective on Chilled Water Systems; PE (HPAC Interactive

www.hpac.com/member/basics/chilled/9808coad.html).

Appendices 285

Bitondo, M.J., Tozzi, M.J. Chiller Plant Control: Multiple Chiller Controls. Carrier

Corporation, 1999.

Chiller-Plant Design in a Deregulated Electric Environment, York International

Corporation (HVAC&R Update).

Chiller Plant Design, Application Guide AG 31-003. McQuay Air-Conditioning, 2001.

Thermal Storage

Bahnfleth, W.P. Cool Thermal Storage: Is It Still Cool? Heating/Piping/AirConditioning.

April, 2002.

Wulfinghoff, D.R. Measure 2.11.1 Install Cooling Thermal Storage, Energy Efficiency

Manual. 1999.

HVAC: Cool Thermal Storage, Los Angeles Department of Water and Power Energy

Advisor Article (www.ladwp.com).

Hot Technology for Large Cooling Systems, Penn State College of Engineering

(www.engr.psu.edu/news/publications).

Real-Time Pricing and Thermal Energy Storage. EPRI HVAC&R Center Quarterly

Newsletter. Winter, 2000.

Silvetti, B.M. The Application of Thermal Storage in an Unregulated Power Marketplace

(www.aeecenter.org/conferencing/backissues/papersil.htm).

Noise and Vibration

Harris, C.M. Noise Control in Buildings. McGraw-Hill, Inc. 1994.

Schaffer, M.E. A Practical Guide to Noise and Vibration Control for HVAC Systems.

ASHRAE, 1991.

Chiller Installation, Operation, and Maintenance

Crowther, H.P. Installing Absorption Chillers. ASHRAE J. July 2000.

Dorgan, P.E., Dorgan, C.E. ASHRAE’s New Chiller Heat Recovery Application Guide.

ASHRAE Transaction. V., 106, Pt 1, 2000.

Julian, C. A Screw Compressor and Chiller Service Plan. Contracting Business Interactive

(www.contracting business.com).

Kao, J.Y. Evaluation of GSA Maintenance Practices of Large Centrifugal Chillers and

Review of GSA Refrigerant Management Practices, National Institute of Standards

and Technology, NISTIR Publication 5336, January 1994.

COOLING TOWERSCooling Tower General References

ASHRAE Handbook, Heating, Ventilating, and Air-Conditionings Systems and

Equipment. ASHRAE, 2000, Chap. 36

Hill, G.B., et al. Cooling Towers: Principles and Practice. 3rd Ed. Butterworth-

Heinemann, 1990.

Appendices286

Davis, D. Cooling Tower Doctor. 1999 (www.ctdoc.com).

Pannkoke, T. Cooling Tower Basics. Heating/Piping/AirConditioning February 1996.

Willa, J.L. Evolution of the Cooling Tower, Paper No. TP91-01. Cooling Technology

Institute, 1991.

Katzel, J. Trends in Cooling Towers. Plant Engineering, January 2000.

Kuharic, I.F. Psychrometrics and the Psychrometer, Paper No. TP81-13. Cooling

Technology Institute, 1981.

Hensley, J.C. Cooling Tower Fundamentals, The Marley Cooling Tower Company. 2nd

Ed. 1985.

Buger, R. Cooling Tower Technology: Maintenance, Upgrading, and Rebuilding. 3rd Ed.

The Fairmont Press, 1994.

Cheremisinoff, N.P., Cheremisinoff, P.N. Cooling Towers: Selection, Design, and

Practice. SciTech Publishers, Inc., 1989.

Sellers, D. Commissioning Cooling Towers. HPAC Engineering February 2002.

Tower Placement and Layout

Cooling Tower Layout, Electronic Library Documents Nos. 1071, 1074, and 1103.

Baltimore Aircoil Company, Baltimore, MD, 1999.

Equipment Layout Manual, Bulletin 311G. EVAPCO, Inc., Westminster, MD, 1999.

HVAC Systems References

Stanford, H.W. II, Analysis and Design of Heating, Ventilating, and Air-Conditioning

Systems. I. Prentice-Hall, 1988 (ISBN 0-13-033044-2) 2nd Ed, 2000, Unpublished.

Noise and Vibration

Harris, C.M. Noise Control in Buildings. McGraw-Hill, Inc., 1994.

Schaffer, M.E. A Practical Guide to Noise and Vibration Control for HVAC Systems.

ASHRAE, 1991.

Gerglund, B., et al. Eds. Guidelines for Community Noise. World Health Organization,

1999.

Controls

Butterfly Valves. Engineering Data Book, Section Vb2. Johnson Controls, Inc., 1990.

Engineering Manual of Automatic Control for Commercial Buildings. I-P Ed. Honeywell,

Inc., 1997 (LC No. 97-072971).

Water Treatment

Industrial Water Conditioning. 9th Ed. Betz Dearborn, 1991.

Meitz, A. Water Treatment for Cooling Towers. Heating/Piping/Air Conditioning

January:125–134, 1999.

Appendices 287

Fundamentals of Cooling Water Treatment. Annual Conference 2000 Panel Discussion,

Cooling Technology Institute. CTI Journal 22(1):56–74.

Ruisinger, T.P. Ozonation in Cooling Water Systems. Plant Engineering August, 1996.

Smithee, B. Cooling Tower Maintenance Improves with Ozone. Facilities Engineering

Journal July/August:2000.

Fire Protection

Standard on Water-Cooling Towers, NFPA 214, National Fire Protection Association,

Boston, 2000.

Appendices288

Index

Absorber, 20

Absorption (see Refrigeration, see Water

chillers)

Adjustable speed drives (see Variable

frequency drives)

Air-Conditioning and Refrigeration

Institute (ARI)

Standard 530/590–98, 37

Standard 570, 88

American Society of Heating,

Refrigerating, and Air-

Conditioning Engineers

(ASHRAE)

Standard 15, 37

Standard 34–1989, 6

Standard 90.1–1999, 38

American Society of Mechanical

Engineers (ASME)

Boiler and Pressure Vessel Code, 37

Ammonia (see Refrigerants)

Chilled water

approach, 9

brine, 10

expansion and air removal,

50–52

air cushion expansion tank, 50, 51

inline tangential air separator, 50, 53

flow rate, 8

HVAC applications, 6–8

maintenance, 51

piping, 48–50

bridge, 28

copper, 49

ductile/cast iron, 49

insulation, 49–50

PVC/CPVC, 49

steel, 48

pumps (see Pumps)

range, 8, 10

storage (see Thermal storage)

supply temperature, 8–10

apparatus dewpoint temperature

(ADP), 9

system configuration, 24–32

289

parallel chillers, 26–32

primary– secondary distribution,

28–29

series chillers, 25–26

single chiller, 24

variable primary distribution, 30–32

water treatment (see Water treatment)

Chiller (see Water chiller)

Coefficient of performance (COP), 15,

17, 20, 22

Compression(seeVaporcompressioncycle)

Compressor (seeRefrigeration, compressor)

Condenser, 2, 3, 116

Condenser water

flow rate, 115, 124, 125

heat rejection, 115, 154–155

piping, 116, 166–168

pumps (see Pumps)

range, 115, 119, 125, 153–154

water treatment (see Water treatment)

Control

chilled water system, 62–70

capacity, 63–65

refrigerant flow, 63

sequencingmultiple chillers, 65, 67

chiller start/stop, 62–63

constant flow, 65–66

pump operation, 68–70

supply water temperature reset, 67

variable flow, 66067, 68–70

cooling coil heat transfer, 68–69

equal percentage control valve, 69

logarithmic mean temperature

difference (LMTD), 68

pump speed control, 70

variable-primaryminimumflow, 70

condenser water system

capacity, 183–190

fan cycling, 184–186

fan speed control, 186–187

tower staging, 189–190

cooling tower start/stop, 179–183

bypass valve, 180–183

pump operation, 179

make-up water

electronic level control, 190

mechanical float valve, 190

safety vibration switch, 190

Cooler (see Evaporator)

Cooling load, 33–36

load profile, 34–36

peak load, 33

Cooling Technology Institute (CTI)

performance guarantee, 240, 241

Standard 114, 132

Standard 119, 132

Standard 201, 240

Cooling towers

application, 155–158

multiple towers or cells, x155–158

single tower, 155

approach, 118

basin, 117, 136

“bottle” cooling tower, 135

capacity, 120–122

casing, 135

configuration, 125–128

counterflow, 127, 147, 152

counterflow vs. crossflow, 150–151

crossflow, 127, 147, 152

venturi, 127–128

control (see Controls)

draft

mechanical, 118, 151–152

forced, 118, 127, 147, 152

induced, 118, 127, 147, 152

drift eliminators, 117, 136

drives (see Cooling towers,

mechanical drives)

elements, 116–118

economizer cycle (see Waterside

economizer cycle)

fan(s), 117, 18, 137–142

axial propeller, 137

centrifugal, 137

backward-inclined blades, 137

forward-curved, blades, 137

performance, 137–138

Index290

fan/system relationship, 138–142

fan laws, 141–142

fill, 116, 129–132

film, 130–132

splash, 130–131

spray, 129–130

fire protection, 223–224

sprinkler systems, 223

water flow rates, 223

heat transfer, 120–122

Merkel equation, 120

tower characteristic, 120, 121

maintenance and operation, 225–237

commissioning, 225–227

heat exchanger, 230–231

mechanical maintenance, 228–230

venturi tower, 230

mechanical drives, 144–146

belt, 144–145

gear, 145–146

noise and vibration, 216–221

community noise, 216

manufacturers’ sound data, 217–218

property line sound limits, 216

World Health Organization (WHO)

sound limits, 216–217, 219

nomenclature, 118–119

performancefactors,122–125,152–155,

233–234

enhancing, 233–234

troubleshooting, 231–233

performance testing, 247–253

in-situ testing, 247–248

preparation for testing, 251–253

testing criteria andmethods, 248–251

piping

condenser water (see Condenser

water, piping)

drain and overflow, 170

make-up water, 168–170

blowdown loss, 169

drift loss, 169

evaporation loss, 169

multiple towers, 170–171

common basin, 172

equalizer line, 172–173

placementandinstallation,86,158–166

adjacency, 159–160

dunnage, 164

elevation, 158, 160

evase discharge, 160–162

grillage, 160, 164

prevailing wind, 162–164

recirculation, 158

roof, 165

screening, 160, 164

plume control, 221–222

purchasing, 238–246

defining requirements, 238–239

economics, 240–245

annual recurring costs, 242–245

first costs, 242

nonrecurring repair and

replacement costs, 245

totalowningandoperatingcost,245

specifications, 245, App. B

range (see Condenser water)

selection and specification

structural frame, 132–135

concrete, 134–135

fiberglass, 135

steel, 133–134

galvanized, 133–134

stainless, 134

wood, 132

sump, 136, 236

water distribution (see wet deck)

water treatment (see Condenser water)

wet deck, 117, 135–136

winter operation, 234–237

freeze protection, 212–213

shutdown, 235–237

District heating systems, 44

Draft (see Cooling towers)

Electric rates,

consumption, 42

demand, 42

demand charge, 42

Index 291

load factor, 43

off-peak, 42, 43

on-peak, 42, 43

ratchet clauses, 43

real time pricing, 76

Expansion valve, 2

Evaporative condensers

and coolers, 176–178

Evaporator, 2

Fuel oil, 43

Heat

gain, 7

latent, 7, 8

of compression, 8

recovery (see Heat recovery)

sensible, 7, 8

total, 8

Motors

electrical service, 88–92

energy-efficient, 143

hermetic, 91

open, 91

cooling, 91–92

open drip proof, 142

standard, 143

starting, 89–91

across-the-line, 89

current, 89, 90, 91

full voltage, 89

maximum allowable

starts, 143

part-winding, 90

solid-state, 91–92

speed, 58, 60, 144

wye-delta, 90–91

totally enclosed air over (TEAO),

142–143

totally enclosed fan cooled (TEFC),

142–143

two-speed, 143–144

dual winding, 144

“pony” motor, 144

single winding, 144

National Fire Protection Association

(NFPA)

Standard 13, 223

Standard 214, 132

Pumps

base-mounted, 57

cavitation, 175

net positive suction head (NPSH),

175–176

centrifugal, 54

double suction, 56

horizontal split case, 56

impeller, 54

line-mounted, 57

vertical split case, 57

head, 57–58, 173–176

friction head, 57–58

residual head, 174

static head, 173–174

horsepower, 58

lift, 173–174

Purge, 15, 20

noncondensables, 15

pump, 20

units, 15

R-11 (see Refrigerants, chlorofluorocar-

bons)


bons)

R-22 (see Refrigerants, hydrochlorofluor-

ocarbons)

R-123 (see Refrigerants, hydrochloro-

fluorocarbons)

R-134A (see Refrigerants, hydrofluoro-

carbons)

R-407C (see Refrigerants, hydrofluoro-

carbons)

R-410A (see Refrigerants, hydrofluoro-

carbons)


bons)

R-717 (see Refrigerants, ammonia)

R-718 (see Refrigerants, water)

References and resources, App. C

Index292

Refrigerants, 4–6

ammonia, 6

chlorofluorocarbons (CFCs), 4

R-11, 4

R-12, 4

R-503, 4

flammability, 6

hydrochlorofluorocarbons (HCFCs), 5

phaseout, 5–6

R-22, 5, 6

avoiding use of, 6

R-123, 5, 6

avoiding use of, 6

hydrofluorocarbons (HFCs), 5

R-134A, 5, 6

R-407C, 5, 6

R-410A, 5, 6

management program, 105–108

certification

equipment, 106

technician, 106

Clean Air Act of 1990, 105

disposal, 107

EPA Action Guide, 108

venting and leaks, 106

primary, 4

safety groups, 6

secondary, 4

toxicity, 6

water, 4

Refrigeration, 1

absorption cycle, 20

compressor, 2

centrifugal, 12–14

capacity control, 12, 64–65

COP, 15

impeller, 13–14

vanes, 12, 64

electric-drive, 15–17

COP, 15

part load performance, 16

engine-drive, 17–18

COP, 17

lean burn technology, 18

life, 18

maintenance cost, 18

natural gas, 17, 18

reciprocating, 11–12

capacity control, 11, 63–64

COP, 15

rotary, 12–15

COP, 15

part load performance, 16

screw, 12, 64

capacity control, 12, 64

condenser, 2

air-cooled, 11, 19, 45

water-cooled, 11

fouling factor, 37, App. B

direct expansion (DX), 6

evaporator, 2, 37

fouling factor, 37, App. B

lift, 13

machinery room, 46

negative pressure systems, 5

purge (see Purge)

positive pressure systems, 5, 14–15

vapor compression cycle, 1–4

compression, 1

condensation, 1

evaporation, 1

expansion, 1

pressure-enthalpy chart, 2–4

Sensible heat ratio (SHR), 9

Thermal storage, 71–85


chilled water storage, 76

tanks, 78

stratification, 78

comparison, 76, 77

economics, 71–76

charging, 72

discharging, 72

operating strategies, 72–76

storage efficiency, 72

eutectic salts storage, 83–84

ice storage, 79–83

external melt ice-on-coil, 80

Index 293

ice capsules, 83

ice shedders, 80

ice slurry, 83

internal melt ice-on-coil, 80

phase change, 79

Underwriters Laboratories (UL)

Standard 465, 37

Vapor compression cycle (see

Refrigeration)

Variable frequency drives, 58–61

bypass switching, 61

efficiency, 61

inverter, 60

main power disconnect, 61

motor heat, 144

pulse width modulation (PWM), 60

rectifier, 60

Variable speed drives (see Variable

frequency drives)

Waste heat, 44

Water (see Refrigerants)

Water chiller, 10

absorption, 19–23

absorber, 20


COP, 20, 22

concentrator, 20

direct-fired, 20, 22

indirect-fired, 20, 22

maintenance, 102–103

purge pump, 20

single-stage, 20–21

two-stage, 20

air-cooled, 19

condensing medium, 19

dual compressor, 17

efficiency and energy cost, 18

electric-drive (see Refrigeration,

compressor)

energy cost, 18

engine-drive (see Refrigeration,

compressor)

heat recovery, 92–94

condenser heat, 93–94

internal source heat pump, 92–93

maintenance and operation, 95–108

commissioning, 95–97

maintenance

absorption chiller, 102–103

checklists, 96–97, 100–103

continual monitoring, 98–99

engine-driven, 102

extended maintenance, 101

periodic checks, 100

program, 98

regularly scheduled maintenance,

100–101

operationing goals, 95

noise and vibration, 46, 86–88

A-Scale sound level, 87–88

community noise, 45

noise, 86

sound, 86

sound pressure level, 87

placement and installation, 45–47

air-cooled, 45

chiller room (see Refrigeration,

machinery room)

recirculation, 45

screening, 45

service access, 45–46

water-cooled, 46–47

purchasing, 109–114, App. B

defining requirements, 109–110

economics, 111–114

annual recurring costs, 113

first costs, 112

nonrecurring repair and

replacement costs, 114

total owning and operating cost,

111, 114

specifications, 114, App. B

ratings, 37–39

integrated part load value (IPLV),

37–38

nonstandard part load value

(NPLV), 38

selection, 37–41

Index294

capacity

asymmetrical, 40–41

load vs. capacity, 39

symmetrical, 40

compressor type, 39–40

condensing medium, 40

mixed energy sources, 41–44

troubleshooting, 103–105

vapor compression, 10–19

water-cooled, 19

Water treatment

chilled water, 52–54

condenser water

biological fouling, 201–204

biocides, 201–201

biofilm, 201

Legionella, 202–203

microbiologically induced

corrosion (MIC), 204

chemical storage and safety,

209–211

safety showers and eye wash

stations, 210

spill control, 209–210

control systems, 204–206

corrosion

galvanic, 197

general, 197

inhibitors, 198–199

localized, 197–198

deposition, 191–197

alkalinity, 193

blowdown, 192

dissolved solids (total), 191–192,

193

hardness, 191

cycles of concentration, 195–196

Langelier Saturation Index (LSI),

193–194

pH, 193

phosphonates, 197

Ryzner Stability Index (RSI),

194

temperature, 192

magnets, 208

ozone, 207

program

contracting, 245–246

management, 227–228

sidestream filtration, 206

UV, 208

white rust, 199–201

wood towers, 208–209

Waterside economizer cycle, 213–215

application, 213, 215

heat exchanger

plate and frame, 215

shell and tube, 214–215

Wet bulb temperature (ambient), 119,

124, 154, App. A

White rust (see Water treatment,

condenser water)

Index 295

HVAC Water Chillers HVAC Water...HVAC Water Chillers and Cooling Towers Fundamentals, Application,...

Documents

Transcript of HVAC Water Chillers HVAC Water...HVAC Water Chillers and Cooling Towers Fundamentals, Application,...